ee0d80c219
Model: fpadovani/eus-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
7112 lines
193 KiB
JSON
7112 lines
193 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 3.0068758057584875,
|
|
"eval_steps": 500,
|
|
"global_step": 3500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 7.491108751296997,
|
|
"epoch": 0.004297378599054577,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 7.3189,
|
|
"mean_token_accuracy": 0.09141870960593224,
|
|
"num_tokens": 10107.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 7.515214681625366,
|
|
"epoch": 0.008594757198109154,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 7.3203,
|
|
"mean_token_accuracy": 0.09664802104234696,
|
|
"num_tokens": 18391.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 7.51798095703125,
|
|
"epoch": 0.01289213579716373,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 7e-06,
|
|
"loss": 7.3336,
|
|
"mean_token_accuracy": 0.09290757700800896,
|
|
"num_tokens": 27061.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 7.496134233474732,
|
|
"epoch": 0.017189514396218308,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 7.3792,
|
|
"mean_token_accuracy": 0.10020544826984405,
|
|
"num_tokens": 36339.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 7.491674518585205,
|
|
"epoch": 0.021486892995272882,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 7.3608,
|
|
"mean_token_accuracy": 0.09223981052637101,
|
|
"num_tokens": 45770.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 7.550447177886963,
|
|
"epoch": 0.02578427159432746,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 7.3766,
|
|
"mean_token_accuracy": 0.09226111099123954,
|
|
"num_tokens": 54575.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 7.5649909496307375,
|
|
"epoch": 0.030081650193382038,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 7.4503,
|
|
"mean_token_accuracy": 0.09087274968624115,
|
|
"num_tokens": 66403.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 7.568733358383179,
|
|
"epoch": 0.034379028792436615,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 7.3758,
|
|
"mean_token_accuracy": 0.09583879709243774,
|
|
"num_tokens": 76510.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 7.536895418167115,
|
|
"epoch": 0.03867640739149119,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 7.2707,
|
|
"mean_token_accuracy": 0.09973574429750443,
|
|
"num_tokens": 84836.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 7.539294195175171,
|
|
"epoch": 0.042973785990545764,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 7.2456,
|
|
"mean_token_accuracy": 0.10200604945421218,
|
|
"num_tokens": 93197.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 7.543957090377807,
|
|
"epoch": 0.047271164589600345,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 7.245,
|
|
"mean_token_accuracy": 0.09889603033661842,
|
|
"num_tokens": 101546.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 7.518790864944458,
|
|
"epoch": 0.05156854318865492,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 7.3185,
|
|
"mean_token_accuracy": 0.0964392140507698,
|
|
"num_tokens": 111703.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 7.569249010086059,
|
|
"epoch": 0.055865921787709494,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 7.2462,
|
|
"mean_token_accuracy": 0.10106701627373696,
|
|
"num_tokens": 119894.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 7.529278564453125,
|
|
"epoch": 0.060163300386764075,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 7.212,
|
|
"mean_token_accuracy": 0.10316545292735099,
|
|
"num_tokens": 128885.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 7.5085443496704105,
|
|
"epoch": 0.06446067898581866,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 7.3341,
|
|
"mean_token_accuracy": 0.09605684131383896,
|
|
"num_tokens": 138106.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 7.545536041259766,
|
|
"epoch": 0.06875805758487323,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 7.2453,
|
|
"mean_token_accuracy": 0.1033376321196556,
|
|
"num_tokens": 146691.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 7.557823944091797,
|
|
"epoch": 0.0730554361839278,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 7.2809,
|
|
"mean_token_accuracy": 0.10218598842620849,
|
|
"num_tokens": 155792.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 7.59003210067749,
|
|
"epoch": 0.07735281478298238,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 7.358,
|
|
"mean_token_accuracy": 0.09359859973192215,
|
|
"num_tokens": 166944.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 7.617101383209229,
|
|
"epoch": 0.08165019338203695,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 7.1989,
|
|
"mean_token_accuracy": 0.1046324297785759,
|
|
"num_tokens": 175303.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 7.528326416015625,
|
|
"epoch": 0.08594757198109153,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 7.2591,
|
|
"mean_token_accuracy": 0.10362932458519936,
|
|
"num_tokens": 184708.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 7.538442516326905,
|
|
"epoch": 0.09024495058014612,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 7.2175,
|
|
"mean_token_accuracy": 0.10121936872601509,
|
|
"num_tokens": 193835.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 7.487286758422852,
|
|
"epoch": 0.09454232917920069,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 7.2608,
|
|
"mean_token_accuracy": 0.10149957984685898,
|
|
"num_tokens": 203344.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 7.519776153564453,
|
|
"epoch": 0.09883970777825526,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 7.365,
|
|
"mean_token_accuracy": 0.090200637280941,
|
|
"num_tokens": 213048.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 7.641110944747925,
|
|
"epoch": 0.10313708637730984,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 7.2283,
|
|
"mean_token_accuracy": 0.10701292306184769,
|
|
"num_tokens": 221784.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 7.459906339645386,
|
|
"epoch": 0.10743446497636441,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 7.2232,
|
|
"mean_token_accuracy": 0.0979800947010517,
|
|
"num_tokens": 230971.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 7.543297910690308,
|
|
"epoch": 0.11173184357541899,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.1925,
|
|
"mean_token_accuracy": 0.10740587189793586,
|
|
"num_tokens": 240524.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 7.5613665103912355,
|
|
"epoch": 0.11602922217447358,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.1644,
|
|
"mean_token_accuracy": 0.10438360348343849,
|
|
"num_tokens": 249220.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 7.50055251121521,
|
|
"epoch": 0.12032660077352815,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.2801,
|
|
"mean_token_accuracy": 0.10291781276464462,
|
|
"num_tokens": 258934.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 7.580437517166137,
|
|
"epoch": 0.12462397937258272,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.1981,
|
|
"mean_token_accuracy": 0.10456435978412629,
|
|
"num_tokens": 267680.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 7.493255138397217,
|
|
"epoch": 0.1289213579716373,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.2271,
|
|
"mean_token_accuracy": 0.10516949817538261,
|
|
"num_tokens": 276227.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 7.456282997131348,
|
|
"epoch": 0.1332187365706919,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.1356,
|
|
"mean_token_accuracy": 0.10240317732095719,
|
|
"num_tokens": 286342.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 7.478529405593872,
|
|
"epoch": 0.13751611516974646,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.2034,
|
|
"mean_token_accuracy": 0.10278507694602013,
|
|
"num_tokens": 294994.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 7.560914468765259,
|
|
"epoch": 0.14181349376880104,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.1728,
|
|
"mean_token_accuracy": 0.10567683354020119,
|
|
"num_tokens": 303882.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 7.53267731666565,
|
|
"epoch": 0.1461108723678556,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.2185,
|
|
"mean_token_accuracy": 0.10334216356277466,
|
|
"num_tokens": 312515.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 7.510133743286133,
|
|
"epoch": 0.15040825096691018,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.1819,
|
|
"mean_token_accuracy": 0.1053494967520237,
|
|
"num_tokens": 320801.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 7.564500951766968,
|
|
"epoch": 0.15470562956596476,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.1958,
|
|
"mean_token_accuracy": 0.10476180166006088,
|
|
"num_tokens": 329382.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 7.523471403121948,
|
|
"epoch": 0.15900300816501933,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.1811,
|
|
"mean_token_accuracy": 0.1035641834139824,
|
|
"num_tokens": 337894.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.490967750549316,
|
|
"epoch": 0.1633003867640739,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.1711,
|
|
"mean_token_accuracy": 0.10333409234881401,
|
|
"num_tokens": 346380.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.575121831893921,
|
|
"epoch": 0.16759776536312848,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.2686,
|
|
"mean_token_accuracy": 0.09736321717500687,
|
|
"num_tokens": 356305.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.543933534622193,
|
|
"epoch": 0.17189514396218306,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.2081,
|
|
"mean_token_accuracy": 0.10070987790822983,
|
|
"num_tokens": 364899.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.537495565414429,
|
|
"epoch": 0.17619252256123766,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.1983,
|
|
"mean_token_accuracy": 0.10088053569197655,
|
|
"num_tokens": 373663.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.536725664138794,
|
|
"epoch": 0.18048990116029223,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.1549,
|
|
"mean_token_accuracy": 0.10402223244309425,
|
|
"num_tokens": 382730.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.5419622421264645,
|
|
"epoch": 0.1847872797593468,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.2121,
|
|
"mean_token_accuracy": 0.09917423650622367,
|
|
"num_tokens": 392676.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.498674964904785,
|
|
"epoch": 0.18908465835840138,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.1809,
|
|
"mean_token_accuracy": 0.09803537428379058,
|
|
"num_tokens": 401050.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.47620153427124,
|
|
"epoch": 0.19338203695745596,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.1693,
|
|
"mean_token_accuracy": 0.10769422501325607,
|
|
"num_tokens": 410009.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.477664566040039,
|
|
"epoch": 0.19767941555651053,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.1813,
|
|
"mean_token_accuracy": 0.1022846557199955,
|
|
"num_tokens": 419302.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.504415702819824,
|
|
"epoch": 0.2019767941555651,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.0897,
|
|
"mean_token_accuracy": 0.11041160002350807,
|
|
"num_tokens": 427296.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.440302228927612,
|
|
"epoch": 0.20627417275461968,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.2172,
|
|
"mean_token_accuracy": 0.10320142880082131,
|
|
"num_tokens": 436368.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.594078588485718,
|
|
"epoch": 0.21057155135367425,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.1116,
|
|
"mean_token_accuracy": 0.10851684138178826,
|
|
"num_tokens": 445535.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.354097700119018,
|
|
"epoch": 0.21486892995272883,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.1184,
|
|
"mean_token_accuracy": 0.10696499496698379,
|
|
"num_tokens": 454769.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.493505668640137,
|
|
"epoch": 0.2191663085517834,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.1528,
|
|
"mean_token_accuracy": 0.10274568796157837,
|
|
"num_tokens": 463975.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.474530839920044,
|
|
"epoch": 0.22346368715083798,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.1706,
|
|
"mean_token_accuracy": 0.10459161475300789,
|
|
"num_tokens": 472899.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.493699789047241,
|
|
"epoch": 0.22776106574989258,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.1087,
|
|
"mean_token_accuracy": 0.10158660039305686,
|
|
"num_tokens": 481556.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.483793687820435,
|
|
"epoch": 0.23205844434894715,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.1274,
|
|
"mean_token_accuracy": 0.10895142033696174,
|
|
"num_tokens": 490253.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.462975406646729,
|
|
"epoch": 0.23635582294800173,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.1663,
|
|
"mean_token_accuracy": 0.10515937954187393,
|
|
"num_tokens": 498444.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.426873064041137,
|
|
"epoch": 0.2406532015470563,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.093,
|
|
"mean_token_accuracy": 0.11008984223008156,
|
|
"num_tokens": 508330.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.484303283691406,
|
|
"epoch": 0.24495058014611087,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.2074,
|
|
"mean_token_accuracy": 0.10344811007380486,
|
|
"num_tokens": 517900.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.455639600753784,
|
|
"epoch": 0.24924795874516545,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.1342,
|
|
"mean_token_accuracy": 0.10256287604570388,
|
|
"num_tokens": 527808.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.514905118942261,
|
|
"epoch": 0.25354533734422,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.1506,
|
|
"mean_token_accuracy": 0.10260518342256546,
|
|
"num_tokens": 536931.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.504227924346924,
|
|
"epoch": 0.2578427159432746,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.1732,
|
|
"mean_token_accuracy": 0.10155779272317886,
|
|
"num_tokens": 545758.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.49675178527832,
|
|
"epoch": 0.26214009454232917,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.1899,
|
|
"mean_token_accuracy": 0.10567427054047585,
|
|
"num_tokens": 555165.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.394629907608032,
|
|
"epoch": 0.2664374731413838,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.0983,
|
|
"mean_token_accuracy": 0.10148947015404701,
|
|
"num_tokens": 564719.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.432728815078735,
|
|
"epoch": 0.2707348517404383,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.0682,
|
|
"mean_token_accuracy": 0.10739247426390648,
|
|
"num_tokens": 573572.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.586349010467529,
|
|
"epoch": 0.2750322303394929,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.2456,
|
|
"mean_token_accuracy": 0.10255748108029365,
|
|
"num_tokens": 581497.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.318369674682617,
|
|
"epoch": 0.27932960893854747,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.1311,
|
|
"mean_token_accuracy": 0.10708568394184112,
|
|
"num_tokens": 591107.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.487468528747558,
|
|
"epoch": 0.28362698753760207,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.1639,
|
|
"mean_token_accuracy": 0.10406824648380279,
|
|
"num_tokens": 600241.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.458619403839111,
|
|
"epoch": 0.2879243661366566,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.0833,
|
|
"mean_token_accuracy": 0.10933952927589416,
|
|
"num_tokens": 608697.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.387399578094483,
|
|
"epoch": 0.2922217447357112,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.0767,
|
|
"mean_token_accuracy": 0.11132601723074913,
|
|
"num_tokens": 617275.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.476171684265137,
|
|
"epoch": 0.29651912333476577,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.1697,
|
|
"mean_token_accuracy": 0.1055254265666008,
|
|
"num_tokens": 626644.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.3781880855560305,
|
|
"epoch": 0.30081650193382037,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.067,
|
|
"mean_token_accuracy": 0.10572395622730255,
|
|
"num_tokens": 635110.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.468985176086425,
|
|
"epoch": 0.30511388053287497,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.2351,
|
|
"mean_token_accuracy": 0.10162713378667831,
|
|
"num_tokens": 644746.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.523141098022461,
|
|
"epoch": 0.3094112591319295,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.1415,
|
|
"mean_token_accuracy": 0.10366995185613632,
|
|
"num_tokens": 654281.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.358317518234253,
|
|
"epoch": 0.3137086377309841,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.1342,
|
|
"mean_token_accuracy": 0.1051877960562706,
|
|
"num_tokens": 663174.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.420715570449829,
|
|
"epoch": 0.31800601633003867,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.0567,
|
|
"mean_token_accuracy": 0.1060252383351326,
|
|
"num_tokens": 672178.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.440627765655518,
|
|
"epoch": 0.32230339492909327,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.0829,
|
|
"mean_token_accuracy": 0.10851024910807609,
|
|
"num_tokens": 681323.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.312611055374146,
|
|
"epoch": 0.3266007735281478,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.0134,
|
|
"mean_token_accuracy": 0.1070731021463871,
|
|
"num_tokens": 690461.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.485116195678711,
|
|
"epoch": 0.3308981521272024,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.038,
|
|
"mean_token_accuracy": 0.10665726512670518,
|
|
"num_tokens": 699199.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.286711549758911,
|
|
"epoch": 0.33519553072625696,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.0332,
|
|
"mean_token_accuracy": 0.11690814569592475,
|
|
"num_tokens": 707949.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.411632871627807,
|
|
"epoch": 0.33949290932531156,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.0682,
|
|
"mean_token_accuracy": 0.10877912119030952,
|
|
"num_tokens": 715752.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.451524162292481,
|
|
"epoch": 0.3437902879243661,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.0304,
|
|
"mean_token_accuracy": 0.11193935424089432,
|
|
"num_tokens": 724416.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.3398294925689695,
|
|
"epoch": 0.3480876665234207,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.0979,
|
|
"mean_token_accuracy": 0.10706080496311188,
|
|
"num_tokens": 733116.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.373892545700073,
|
|
"epoch": 0.3523850451224753,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.0823,
|
|
"mean_token_accuracy": 0.10571770668029785,
|
|
"num_tokens": 742093.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.35628867149353,
|
|
"epoch": 0.35668242372152986,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.0485,
|
|
"mean_token_accuracy": 0.11085971966385841,
|
|
"num_tokens": 750402.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.272590160369873,
|
|
"epoch": 0.36097980232058446,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.0175,
|
|
"mean_token_accuracy": 0.11432302817702293,
|
|
"num_tokens": 760961.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.479759454727173,
|
|
"epoch": 0.365277180919639,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.1296,
|
|
"mean_token_accuracy": 0.10654517561197281,
|
|
"num_tokens": 770554.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.259108257293701,
|
|
"epoch": 0.3695745595186936,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.0763,
|
|
"mean_token_accuracy": 0.11016570776700974,
|
|
"num_tokens": 779172.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.46284122467041,
|
|
"epoch": 0.37387193811774816,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.1502,
|
|
"mean_token_accuracy": 0.10204855352640152,
|
|
"num_tokens": 788040.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.377580165863037,
|
|
"epoch": 0.37816931671680276,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.0556,
|
|
"mean_token_accuracy": 0.10719484761357308,
|
|
"num_tokens": 796786.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.2963978290557865,
|
|
"epoch": 0.3824666953158573,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.0257,
|
|
"mean_token_accuracy": 0.11093981638550758,
|
|
"num_tokens": 805520.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.230100870132446,
|
|
"epoch": 0.3867640739149119,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.9865,
|
|
"mean_token_accuracy": 0.10948721244931221,
|
|
"num_tokens": 814939.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.344197130203247,
|
|
"epoch": 0.39106145251396646,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.9985,
|
|
"mean_token_accuracy": 0.11251249313354492,
|
|
"num_tokens": 823862.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.369870948791504,
|
|
"epoch": 0.39535883111302106,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.0169,
|
|
"mean_token_accuracy": 0.10975069999694824,
|
|
"num_tokens": 832820.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.322132015228272,
|
|
"epoch": 0.39965620971207566,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.0044,
|
|
"mean_token_accuracy": 0.11229758933186532,
|
|
"num_tokens": 841538.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.337738561630249,
|
|
"epoch": 0.4039535883111302,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.1229,
|
|
"mean_token_accuracy": 0.10578583925962448,
|
|
"num_tokens": 851123.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.333683729171753,
|
|
"epoch": 0.4082509669101848,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.0822,
|
|
"mean_token_accuracy": 0.1119972050189972,
|
|
"num_tokens": 860357.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.331111764907837,
|
|
"epoch": 0.41254834550923936,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.9903,
|
|
"mean_token_accuracy": 0.11009360924363136,
|
|
"num_tokens": 869980.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.231062650680542,
|
|
"epoch": 0.41684572410829396,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.9683,
|
|
"mean_token_accuracy": 0.11301724538207054,
|
|
"num_tokens": 878250.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.38806791305542,
|
|
"epoch": 0.4211431027073485,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.0717,
|
|
"mean_token_accuracy": 0.10711101293563843,
|
|
"num_tokens": 887624.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.216950559616089,
|
|
"epoch": 0.4254404813064031,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.0432,
|
|
"mean_token_accuracy": 0.11447345688939095,
|
|
"num_tokens": 897120.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.223309516906738,
|
|
"epoch": 0.42973785990545765,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.0009,
|
|
"mean_token_accuracy": 0.11249879151582717,
|
|
"num_tokens": 906215.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.42973785990545765,
|
|
"eval_entropy": 7.237139492421536,
|
|
"eval_loss": 7.02439546585083,
|
|
"eval_mean_token_accuracy": 0.11452774357822565,
|
|
"eval_num_tokens": 906215.0,
|
|
"eval_runtime": 2.6841,
|
|
"eval_samples_per_second": 1322.226,
|
|
"eval_steps_per_second": 165.418,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.341209077835083,
|
|
"epoch": 0.43403523850451226,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.0128,
|
|
"mean_token_accuracy": 0.11152525022625923,
|
|
"num_tokens": 915181.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.27630581855774,
|
|
"epoch": 0.4383326171035668,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.0197,
|
|
"mean_token_accuracy": 0.1118933878839016,
|
|
"num_tokens": 924377.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.359603357315064,
|
|
"epoch": 0.4426299957026214,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.1061,
|
|
"mean_token_accuracy": 0.10802508145570755,
|
|
"num_tokens": 933114.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.359967756271362,
|
|
"epoch": 0.44692737430167595,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.0622,
|
|
"mean_token_accuracy": 0.10966231301426888,
|
|
"num_tokens": 943306.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.2014100551605225,
|
|
"epoch": 0.45122475290073055,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.9652,
|
|
"mean_token_accuracy": 0.1156508482992649,
|
|
"num_tokens": 951515.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.308715724945069,
|
|
"epoch": 0.45552213149978515,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.0944,
|
|
"mean_token_accuracy": 0.10246431827545166,
|
|
"num_tokens": 962686.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.312524938583374,
|
|
"epoch": 0.4598195100988397,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.0842,
|
|
"mean_token_accuracy": 0.11101078242063522,
|
|
"num_tokens": 972136.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.238412475585937,
|
|
"epoch": 0.4641168886978943,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.9806,
|
|
"mean_token_accuracy": 0.11057559326291085,
|
|
"num_tokens": 981301.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.246242666244507,
|
|
"epoch": 0.46841426729694885,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.0099,
|
|
"mean_token_accuracy": 0.10596411451697349,
|
|
"num_tokens": 990360.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.30305585861206,
|
|
"epoch": 0.47271164589600345,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.0565,
|
|
"mean_token_accuracy": 0.10379971489310265,
|
|
"num_tokens": 999415.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.228947544097901,
|
|
"epoch": 0.477009024495058,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.9819,
|
|
"mean_token_accuracy": 0.10675028637051583,
|
|
"num_tokens": 1008762.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.223627710342408,
|
|
"epoch": 0.4813064030941126,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.9712,
|
|
"mean_token_accuracy": 0.11833404675126076,
|
|
"num_tokens": 1017704.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.206258535385132,
|
|
"epoch": 0.48560378169316715,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.9319,
|
|
"mean_token_accuracy": 0.11355781629681587,
|
|
"num_tokens": 1026251.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.172951602935791,
|
|
"epoch": 0.48990116029222175,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.9471,
|
|
"mean_token_accuracy": 0.11119012758135796,
|
|
"num_tokens": 1036191.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.232996034622192,
|
|
"epoch": 0.4941985388912763,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.9737,
|
|
"mean_token_accuracy": 0.11196622401475906,
|
|
"num_tokens": 1044936.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.271992301940918,
|
|
"epoch": 0.4984959174903309,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.9757,
|
|
"mean_token_accuracy": 0.11317787989974022,
|
|
"num_tokens": 1053683.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.157695055007935,
|
|
"epoch": 0.5027932960893855,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.9486,
|
|
"mean_token_accuracy": 0.11400579363107681,
|
|
"num_tokens": 1062932.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.19578332901001,
|
|
"epoch": 0.50709067468844,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.9565,
|
|
"mean_token_accuracy": 0.11118854060769082,
|
|
"num_tokens": 1072313.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.2023499488830565,
|
|
"epoch": 0.5113880532874946,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.9656,
|
|
"mean_token_accuracy": 0.10927657783031464,
|
|
"num_tokens": 1081675.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.180068778991699,
|
|
"epoch": 0.5156854318865493,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.9584,
|
|
"mean_token_accuracy": 0.10982275977730752,
|
|
"num_tokens": 1091541.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.253817129135132,
|
|
"epoch": 0.5199828104856038,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.9824,
|
|
"mean_token_accuracy": 0.11794964224100113,
|
|
"num_tokens": 1100724.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.123758220672608,
|
|
"epoch": 0.5242801890846583,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.9407,
|
|
"mean_token_accuracy": 0.11821772009134293,
|
|
"num_tokens": 1108869.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.2373988151550295,
|
|
"epoch": 0.5285775676837129,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.968,
|
|
"mean_token_accuracy": 0.1125386081635952,
|
|
"num_tokens": 1117314.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.098644876480103,
|
|
"epoch": 0.5328749462827675,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.9317,
|
|
"mean_token_accuracy": 0.11509137749671935,
|
|
"num_tokens": 1126786.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.08965916633606,
|
|
"epoch": 0.5371723248818221,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.9196,
|
|
"mean_token_accuracy": 0.11359397768974304,
|
|
"num_tokens": 1136013.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.193327474594116,
|
|
"epoch": 0.5414697034808766,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.9515,
|
|
"mean_token_accuracy": 0.11913771256804466,
|
|
"num_tokens": 1144970.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.144840240478516,
|
|
"epoch": 0.5457670820799312,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.9041,
|
|
"mean_token_accuracy": 0.12084828615188599,
|
|
"num_tokens": 1153810.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.152580881118775,
|
|
"epoch": 0.5500644606789858,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.9033,
|
|
"mean_token_accuracy": 0.11283695176243783,
|
|
"num_tokens": 1162498.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.078862094879151,
|
|
"epoch": 0.5543618392780404,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.9844,
|
|
"mean_token_accuracy": 0.11267361715435982,
|
|
"num_tokens": 1172091.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.244156694412231,
|
|
"epoch": 0.5586592178770949,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.9711,
|
|
"mean_token_accuracy": 0.1131252221763134,
|
|
"num_tokens": 1181400.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.147411584854126,
|
|
"epoch": 0.5629565964761496,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.9429,
|
|
"mean_token_accuracy": 0.11570452973246574,
|
|
"num_tokens": 1189780.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.069279909133911,
|
|
"epoch": 0.5672539750752041,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.9177,
|
|
"mean_token_accuracy": 0.11673617735505104,
|
|
"num_tokens": 1198671.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.186899328231812,
|
|
"epoch": 0.5715513536742587,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.9346,
|
|
"mean_token_accuracy": 0.11273332387208938,
|
|
"num_tokens": 1207173.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.0509740829467775,
|
|
"epoch": 0.5758487322733132,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.8633,
|
|
"mean_token_accuracy": 0.11901918649673462,
|
|
"num_tokens": 1216387.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.1644703388214115,
|
|
"epoch": 0.5801461108723679,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.861,
|
|
"mean_token_accuracy": 0.11411493569612503,
|
|
"num_tokens": 1224461.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.09780912399292,
|
|
"epoch": 0.5844434894714224,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.9331,
|
|
"mean_token_accuracy": 0.11777294874191284,
|
|
"num_tokens": 1233774.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.0529937744140625,
|
|
"epoch": 0.588740868070477,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.8872,
|
|
"mean_token_accuracy": 0.11746734082698822,
|
|
"num_tokens": 1242812.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.235918474197388,
|
|
"epoch": 0.5930382466695315,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.0025,
|
|
"mean_token_accuracy": 0.10898117944598198,
|
|
"num_tokens": 1252872.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.88110318183899,
|
|
"epoch": 0.5973356252685862,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.7799,
|
|
"mean_token_accuracy": 0.11457612216472626,
|
|
"num_tokens": 1260852.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.073757219314575,
|
|
"epoch": 0.6016330038676407,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.8843,
|
|
"mean_token_accuracy": 0.11701429635286331,
|
|
"num_tokens": 1268925.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.184703063964844,
|
|
"epoch": 0.6059303824666953,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.9396,
|
|
"mean_token_accuracy": 0.10898746475577355,
|
|
"num_tokens": 1278994.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.032673788070679,
|
|
"epoch": 0.6102277610657499,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.9216,
|
|
"mean_token_accuracy": 0.11066435426473617,
|
|
"num_tokens": 1287698.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.109692239761353,
|
|
"epoch": 0.6145251396648045,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.853,
|
|
"mean_token_accuracy": 0.11746916100382805,
|
|
"num_tokens": 1297475.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.045703029632568,
|
|
"epoch": 0.618822518263859,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.8945,
|
|
"mean_token_accuracy": 0.11652352586388588,
|
|
"num_tokens": 1306836.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.154710102081299,
|
|
"epoch": 0.6231198968629136,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.9426,
|
|
"mean_token_accuracy": 0.10900169834494591,
|
|
"num_tokens": 1315872.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.087551069259644,
|
|
"epoch": 0.6274172754619682,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.8936,
|
|
"mean_token_accuracy": 0.1230031467974186,
|
|
"num_tokens": 1324624.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.031505298614502,
|
|
"epoch": 0.6317146540610228,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.8703,
|
|
"mean_token_accuracy": 0.12103518173098564,
|
|
"num_tokens": 1333058.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.979029607772827,
|
|
"epoch": 0.6360120326600773,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.8463,
|
|
"mean_token_accuracy": 0.11296493709087371,
|
|
"num_tokens": 1342376.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.05972146987915,
|
|
"epoch": 0.6403094112591319,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.8815,
|
|
"mean_token_accuracy": 0.10966399237513542,
|
|
"num_tokens": 1351386.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.994411182403565,
|
|
"epoch": 0.6446067898581865,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.7743,
|
|
"mean_token_accuracy": 0.12411837130784989,
|
|
"num_tokens": 1358958.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.1177832126617435,
|
|
"epoch": 0.6489041684572411,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.9423,
|
|
"mean_token_accuracy": 0.1123286470770836,
|
|
"num_tokens": 1368599.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.095702123641968,
|
|
"epoch": 0.6532015470562956,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.9805,
|
|
"mean_token_accuracy": 0.10967258512973785,
|
|
"num_tokens": 1378529.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.926534938812256,
|
|
"epoch": 0.6574989256553503,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.7622,
|
|
"mean_token_accuracy": 0.12447424307465553,
|
|
"num_tokens": 1386993.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.022530889511108,
|
|
"epoch": 0.6617963042544048,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.8542,
|
|
"mean_token_accuracy": 0.1206921823322773,
|
|
"num_tokens": 1395790.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.937219333648682,
|
|
"epoch": 0.6660936828534594,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.8662,
|
|
"mean_token_accuracy": 0.11413793414831161,
|
|
"num_tokens": 1405587.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.998539781570434,
|
|
"epoch": 0.6703910614525139,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.8418,
|
|
"mean_token_accuracy": 0.11912463083863259,
|
|
"num_tokens": 1414478.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.073210048675537,
|
|
"epoch": 0.6746884400515686,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.9072,
|
|
"mean_token_accuracy": 0.11900003850460053,
|
|
"num_tokens": 1423791.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.071209669113159,
|
|
"epoch": 0.6789858186506231,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.9546,
|
|
"mean_token_accuracy": 0.10661662891507148,
|
|
"num_tokens": 1432955.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.010345840454102,
|
|
"epoch": 0.6832831972496777,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.9059,
|
|
"mean_token_accuracy": 0.11467154771089554,
|
|
"num_tokens": 1441907.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.940794563293457,
|
|
"epoch": 0.6875805758487322,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.7741,
|
|
"mean_token_accuracy": 0.12223436459898948,
|
|
"num_tokens": 1451062.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.077725887298584,
|
|
"epoch": 0.6918779544477869,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.8575,
|
|
"mean_token_accuracy": 0.11200195476412773,
|
|
"num_tokens": 1460132.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.851260232925415,
|
|
"epoch": 0.6961753330468414,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.7443,
|
|
"mean_token_accuracy": 0.12217162400484086,
|
|
"num_tokens": 1469582.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.973898649215698,
|
|
"epoch": 0.700472711645896,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.8383,
|
|
"mean_token_accuracy": 0.11406316757202148,
|
|
"num_tokens": 1479053.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.982749509811401,
|
|
"epoch": 0.7047700902449506,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.8506,
|
|
"mean_token_accuracy": 0.1121716171503067,
|
|
"num_tokens": 1488189.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.829049015045166,
|
|
"epoch": 0.7090674688440052,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.7189,
|
|
"mean_token_accuracy": 0.12386269792914391,
|
|
"num_tokens": 1497324.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.92303500175476,
|
|
"epoch": 0.7133648474430597,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.8317,
|
|
"mean_token_accuracy": 0.1222977451980114,
|
|
"num_tokens": 1506543.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.996142625808716,
|
|
"epoch": 0.7176622260421143,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.8217,
|
|
"mean_token_accuracy": 0.1265745148062706,
|
|
"num_tokens": 1516737.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.881067752838135,
|
|
"epoch": 0.7219596046411689,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.6835,
|
|
"mean_token_accuracy": 0.12508728802204133,
|
|
"num_tokens": 1525561.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.863363933563233,
|
|
"epoch": 0.7262569832402235,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.7097,
|
|
"mean_token_accuracy": 0.1254148416221142,
|
|
"num_tokens": 1533323.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.951838445663452,
|
|
"epoch": 0.730554361839278,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.7733,
|
|
"mean_token_accuracy": 0.12557526156306267,
|
|
"num_tokens": 1542632.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.895253658294678,
|
|
"epoch": 0.7348517404383326,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.736,
|
|
"mean_token_accuracy": 0.12572805732488632,
|
|
"num_tokens": 1551236.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.809239673614502,
|
|
"epoch": 0.7391491190373872,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.7388,
|
|
"mean_token_accuracy": 0.11835454404354095,
|
|
"num_tokens": 1559674.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.876753854751587,
|
|
"epoch": 0.7434464976364418,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.8232,
|
|
"mean_token_accuracy": 0.12170301601290703,
|
|
"num_tokens": 1569481.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.016146326065064,
|
|
"epoch": 0.7477438762354963,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.8623,
|
|
"mean_token_accuracy": 0.11574919894337654,
|
|
"num_tokens": 1578488.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.8354627132415775,
|
|
"epoch": 0.752041254834551,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.751,
|
|
"mean_token_accuracy": 0.12034946754574775,
|
|
"num_tokens": 1586675.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.949285078048706,
|
|
"epoch": 0.7563386334336055,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.8707,
|
|
"mean_token_accuracy": 0.11235570311546325,
|
|
"num_tokens": 1595411.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.988899230957031,
|
|
"epoch": 0.7606360120326601,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.8369,
|
|
"mean_token_accuracy": 0.11851750463247299,
|
|
"num_tokens": 1604046.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.915553617477417,
|
|
"epoch": 0.7649333906317146,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.8301,
|
|
"mean_token_accuracy": 0.1143190972507,
|
|
"num_tokens": 1613759.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.809361219406128,
|
|
"epoch": 0.7692307692307693,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.8007,
|
|
"mean_token_accuracy": 0.1207757331430912,
|
|
"num_tokens": 1623323.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.934272289276123,
|
|
"epoch": 0.7735281478298238,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.7856,
|
|
"mean_token_accuracy": 0.11662489771842957,
|
|
"num_tokens": 1631727.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.889858341217041,
|
|
"epoch": 0.7778255264288784,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.703,
|
|
"mean_token_accuracy": 0.12306550592184066,
|
|
"num_tokens": 1639544.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.9098219871521,
|
|
"epoch": 0.7821229050279329,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.8408,
|
|
"mean_token_accuracy": 0.11140461266040802,
|
|
"num_tokens": 1648931.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.821422386169433,
|
|
"epoch": 0.7864202836269876,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.711,
|
|
"mean_token_accuracy": 0.1199341893196106,
|
|
"num_tokens": 1657688.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.889161062240601,
|
|
"epoch": 0.7907176622260421,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.8008,
|
|
"mean_token_accuracy": 0.12064535990357399,
|
|
"num_tokens": 1666879.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.840409898757935,
|
|
"epoch": 0.7950150408250967,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.7666,
|
|
"mean_token_accuracy": 0.12245901748538017,
|
|
"num_tokens": 1676773.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.8722686767578125,
|
|
"epoch": 0.7993124194241513,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.8727,
|
|
"mean_token_accuracy": 0.11736220642924308,
|
|
"num_tokens": 1686144.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.927791166305542,
|
|
"epoch": 0.8036097980232059,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.8363,
|
|
"mean_token_accuracy": 0.11194746792316437,
|
|
"num_tokens": 1695476.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.838814640045166,
|
|
"epoch": 0.8079071766222604,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.7542,
|
|
"mean_token_accuracy": 0.11981326714158058,
|
|
"num_tokens": 1704907.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.837986898422241,
|
|
"epoch": 0.812204555221315,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.7911,
|
|
"mean_token_accuracy": 0.1210789442062378,
|
|
"num_tokens": 1714564.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.931927728652954,
|
|
"epoch": 0.8165019338203696,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.8032,
|
|
"mean_token_accuracy": 0.11883696243166923,
|
|
"num_tokens": 1725285.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.8115462303161625,
|
|
"epoch": 0.8207993124194242,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.7991,
|
|
"mean_token_accuracy": 0.11632968410849572,
|
|
"num_tokens": 1734331.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.857008695602417,
|
|
"epoch": 0.8250966910184787,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.726,
|
|
"mean_token_accuracy": 0.1166811965405941,
|
|
"num_tokens": 1742340.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.806628656387329,
|
|
"epoch": 0.8293940696175333,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.7189,
|
|
"mean_token_accuracy": 0.12429693192243577,
|
|
"num_tokens": 1751725.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.705836391448974,
|
|
"epoch": 0.8336914482165879,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.6562,
|
|
"mean_token_accuracy": 0.12673527225852013,
|
|
"num_tokens": 1760294.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.776405715942383,
|
|
"epoch": 0.8379888268156425,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.7301,
|
|
"mean_token_accuracy": 0.12227423414587975,
|
|
"num_tokens": 1768912.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.929677724838257,
|
|
"epoch": 0.842286205414697,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.9124,
|
|
"mean_token_accuracy": 0.11897997409105301,
|
|
"num_tokens": 1778633.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.877961349487305,
|
|
"epoch": 0.8465835840137517,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.7875,
|
|
"mean_token_accuracy": 0.125636800378561,
|
|
"num_tokens": 1787275.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.902551507949829,
|
|
"epoch": 0.8508809626128062,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.7787,
|
|
"mean_token_accuracy": 0.11841877847909928,
|
|
"num_tokens": 1795994.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.722687721252441,
|
|
"epoch": 0.8551783412118608,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.7049,
|
|
"mean_token_accuracy": 0.12368022575974465,
|
|
"num_tokens": 1806379.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.836816644668579,
|
|
"epoch": 0.8594757198109153,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.7419,
|
|
"mean_token_accuracy": 0.12570958435535431,
|
|
"num_tokens": 1816135.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.8594757198109153,
|
|
"eval_entropy": 6.472366559612858,
|
|
"eval_loss": 6.754819869995117,
|
|
"eval_mean_token_accuracy": 0.12417131071692114,
|
|
"eval_num_tokens": 1816135.0,
|
|
"eval_runtime": 2.6416,
|
|
"eval_samples_per_second": 1343.488,
|
|
"eval_steps_per_second": 168.078,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.819082975387573,
|
|
"epoch": 0.86377309840997,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004999998427807679,
|
|
"loss": 6.7428,
|
|
"mean_token_accuracy": 0.12198886126279831,
|
|
"num_tokens": 1824777.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.709800720214844,
|
|
"epoch": 0.8680704770090245,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004999992040780138,
|
|
"loss": 6.7973,
|
|
"mean_token_accuracy": 0.1257079139351845,
|
|
"num_tokens": 1833807.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.860920858383179,
|
|
"epoch": 0.8723678556080791,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999980740669294,
|
|
"loss": 6.7525,
|
|
"mean_token_accuracy": 0.12449089586734771,
|
|
"num_tokens": 1843375.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.839705562591552,
|
|
"epoch": 0.8766652342071336,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004999964527499823,
|
|
"loss": 6.8141,
|
|
"mean_token_accuracy": 0.11705794557929039,
|
|
"num_tokens": 1853036.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.833159351348877,
|
|
"epoch": 0.8809626128061883,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999943401307127,
|
|
"loss": 6.7628,
|
|
"mean_token_accuracy": 0.11710175424814225,
|
|
"num_tokens": 1862041.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.74225492477417,
|
|
"epoch": 0.8852599914052428,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999917362137337,
|
|
"loss": 6.6787,
|
|
"mean_token_accuracy": 0.12101037502288818,
|
|
"num_tokens": 1870707.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.785744285583496,
|
|
"epoch": 0.8895573700042974,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999886410047312,
|
|
"loss": 6.6862,
|
|
"mean_token_accuracy": 0.12277518585324287,
|
|
"num_tokens": 1879787.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.834198951721191,
|
|
"epoch": 0.8938547486033519,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999850545104638,
|
|
"loss": 6.7421,
|
|
"mean_token_accuracy": 0.12042063474655151,
|
|
"num_tokens": 1889413.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.72258677482605,
|
|
"epoch": 0.8981521272024066,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004999809767387633,
|
|
"loss": 6.7197,
|
|
"mean_token_accuracy": 0.11978086978197097,
|
|
"num_tokens": 1898283.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.7828127384185795,
|
|
"epoch": 0.9024495058014611,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999764076985337,
|
|
"loss": 6.7543,
|
|
"mean_token_accuracy": 0.12366160079836845,
|
|
"num_tokens": 1907175.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.719510412216186,
|
|
"epoch": 0.9067468844005157,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999713473997519,
|
|
"loss": 6.8002,
|
|
"mean_token_accuracy": 0.12121418863534927,
|
|
"num_tokens": 1918223.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.836523628234863,
|
|
"epoch": 0.9110442629995703,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999657958534677,
|
|
"loss": 6.7295,
|
|
"mean_token_accuracy": 0.11709223091602325,
|
|
"num_tokens": 1928801.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.680103588104248,
|
|
"epoch": 0.9153416415986249,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999597530718034,
|
|
"loss": 6.7014,
|
|
"mean_token_accuracy": 0.12364157661795616,
|
|
"num_tokens": 1937406.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.803171920776367,
|
|
"epoch": 0.9196390201976794,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000499953219067954,
|
|
"loss": 6.6924,
|
|
"mean_token_accuracy": 0.12274096831679344,
|
|
"num_tokens": 1947184.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.794332647323609,
|
|
"epoch": 0.923936398796734,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999461938561873,
|
|
"loss": 6.719,
|
|
"mean_token_accuracy": 0.12224036455154419,
|
|
"num_tokens": 1956293.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.758011388778686,
|
|
"epoch": 0.9282337773957886,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999386774518432,
|
|
"loss": 6.7009,
|
|
"mean_token_accuracy": 0.12048266977071762,
|
|
"num_tokens": 1964791.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.697661304473877,
|
|
"epoch": 0.9325311559948432,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999306698713349,
|
|
"loss": 6.6068,
|
|
"mean_token_accuracy": 0.1272837094962597,
|
|
"num_tokens": 1973754.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.7667817115783695,
|
|
"epoch": 0.9368285345938977,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999221711321477,
|
|
"loss": 6.6887,
|
|
"mean_token_accuracy": 0.12394253239035606,
|
|
"num_tokens": 1983035.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.704556512832641,
|
|
"epoch": 0.9411259131929522,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999131812528393,
|
|
"loss": 6.7139,
|
|
"mean_token_accuracy": 0.12714729234576225,
|
|
"num_tokens": 1992584.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.736967897415161,
|
|
"epoch": 0.9454232917920069,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00049990370025304,
|
|
"loss": 6.7395,
|
|
"mean_token_accuracy": 0.1247944861650467,
|
|
"num_tokens": 2001876.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.762415361404419,
|
|
"epoch": 0.9497206703910615,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998937281534526,
|
|
"loss": 6.6302,
|
|
"mean_token_accuracy": 0.13029390051960946,
|
|
"num_tokens": 2011067.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.732195615768433,
|
|
"epoch": 0.954018048990116,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004998832649758521,
|
|
"loss": 6.7277,
|
|
"mean_token_accuracy": 0.11973504275083542,
|
|
"num_tokens": 2020763.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.740197038650512,
|
|
"epoch": 0.9583154275891707,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004998723107430862,
|
|
"loss": 6.6922,
|
|
"mean_token_accuracy": 0.12620774358510972,
|
|
"num_tokens": 2029534.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.712693786621093,
|
|
"epoch": 0.9626128061882252,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004998608654790741,
|
|
"loss": 6.6464,
|
|
"mean_token_accuracy": 0.12175018936395646,
|
|
"num_tokens": 2039143.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.659943866729736,
|
|
"epoch": 0.9669101847872797,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000499848929208808,
|
|
"loss": 6.6024,
|
|
"mean_token_accuracy": 0.13268646001815795,
|
|
"num_tokens": 2048253.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.671807098388672,
|
|
"epoch": 0.9712075633863343,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004998365019583519,
|
|
"loss": 6.6707,
|
|
"mean_token_accuracy": 0.13119613155722618,
|
|
"num_tokens": 2057234.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.749271821975708,
|
|
"epoch": 0.975504941985389,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004998235837548417,
|
|
"loss": 6.7138,
|
|
"mean_token_accuracy": 0.1277630090713501,
|
|
"num_tokens": 2065431.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.724707078933716,
|
|
"epoch": 0.9798023205844435,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000499810174626486,
|
|
"loss": 6.7137,
|
|
"mean_token_accuracy": 0.12455461621284485,
|
|
"num_tokens": 2074723.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.69525032043457,
|
|
"epoch": 0.984099699183498,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004997962746025646,
|
|
"loss": 6.5975,
|
|
"mean_token_accuracy": 0.13215404078364373,
|
|
"num_tokens": 2084509.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.723079824447632,
|
|
"epoch": 0.9883970777825526,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004997818837134298,
|
|
"loss": 6.7185,
|
|
"mean_token_accuracy": 0.12199998944997788,
|
|
"num_tokens": 2093110.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.635384130477905,
|
|
"epoch": 0.9926944563816072,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004997670019905057,
|
|
"loss": 6.5836,
|
|
"mean_token_accuracy": 0.12477853298187255,
|
|
"num_tokens": 2102355.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.679490280151367,
|
|
"epoch": 0.9969918349806618,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004997516294662876,
|
|
"loss": 6.6247,
|
|
"mean_token_accuracy": 0.1251891292631626,
|
|
"num_tokens": 2110418.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.690038098229302,
|
|
"epoch": 1.0008594757198108,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004997357661743433,
|
|
"loss": 6.5994,
|
|
"mean_token_accuracy": 0.13212493724293178,
|
|
"num_tokens": 2117866.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.635151147842407,
|
|
"epoch": 1.0051568543188656,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004997194121493118,
|
|
"loss": 6.4157,
|
|
"mean_token_accuracy": 0.1288985140621662,
|
|
"num_tokens": 2126082.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.5850855827331545,
|
|
"epoch": 1.0094542329179201,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997025674269037,
|
|
"loss": 6.3647,
|
|
"mean_token_accuracy": 0.13999661505222322,
|
|
"num_tokens": 2134042.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.608558988571167,
|
|
"epoch": 1.0137516115169747,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004996852320439013,
|
|
"loss": 6.4719,
|
|
"mean_token_accuracy": 0.13355429843068123,
|
|
"num_tokens": 2142570.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.5898195743560795,
|
|
"epoch": 1.0180489901160292,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004996674060381578,
|
|
"loss": 6.408,
|
|
"mean_token_accuracy": 0.13155465126037597,
|
|
"num_tokens": 2151310.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.65716142654419,
|
|
"epoch": 1.0223463687150838,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004996490894485985,
|
|
"loss": 6.4649,
|
|
"mean_token_accuracy": 0.1299987778067589,
|
|
"num_tokens": 2160662.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.589327764511109,
|
|
"epoch": 1.0266437473141383,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004996302823152193,
|
|
"loss": 6.4217,
|
|
"mean_token_accuracy": 0.1353737846016884,
|
|
"num_tokens": 2170067.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.548572969436646,
|
|
"epoch": 1.0309411259131929,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004996109846790873,
|
|
"loss": 6.3812,
|
|
"mean_token_accuracy": 0.13803449124097825,
|
|
"num_tokens": 2178850.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.5560492992401125,
|
|
"epoch": 1.0352385045122476,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004995911965823412,
|
|
"loss": 6.4036,
|
|
"mean_token_accuracy": 0.13800363689661027,
|
|
"num_tokens": 2188307.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.570348930358887,
|
|
"epoch": 1.0395358831113022,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004995709180681899,
|
|
"loss": 6.4088,
|
|
"mean_token_accuracy": 0.14277965128421782,
|
|
"num_tokens": 2197026.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.5593053817749025,
|
|
"epoch": 1.0438332617103567,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000499550149180914,
|
|
"loss": 6.3881,
|
|
"mean_token_accuracy": 0.1331556722521782,
|
|
"num_tokens": 2205537.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.566903448104858,
|
|
"epoch": 1.0481306403094113,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004995288899658641,
|
|
"loss": 6.3982,
|
|
"mean_token_accuracy": 0.13826436847448348,
|
|
"num_tokens": 2214508.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.582273006439209,
|
|
"epoch": 1.0524280189084658,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004995071404694619,
|
|
"loss": 6.529,
|
|
"mean_token_accuracy": 0.12735923379659653,
|
|
"num_tokens": 2223084.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.660481309890747,
|
|
"epoch": 1.0567253975075204,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004994849007391996,
|
|
"loss": 6.4492,
|
|
"mean_token_accuracy": 0.1277565985918045,
|
|
"num_tokens": 2231406.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.501922702789306,
|
|
"epoch": 1.061022776106575,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004994621708236401,
|
|
"loss": 6.363,
|
|
"mean_token_accuracy": 0.1401769794523716,
|
|
"num_tokens": 2239867.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.563766050338745,
|
|
"epoch": 1.0653201547056295,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000499438950772416,
|
|
"loss": 6.4318,
|
|
"mean_token_accuracy": 0.1355653516948223,
|
|
"num_tokens": 2248844.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.487766885757447,
|
|
"epoch": 1.0696175333046842,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004994152406362311,
|
|
"loss": 6.3553,
|
|
"mean_token_accuracy": 0.14220800772309303,
|
|
"num_tokens": 2257599.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.559006404876709,
|
|
"epoch": 1.0739149119037388,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004993910404668586,
|
|
"loss": 6.3984,
|
|
"mean_token_accuracy": 0.1350975051522255,
|
|
"num_tokens": 2266510.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.513402509689331,
|
|
"epoch": 1.0782122905027933,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000499366350317142,
|
|
"loss": 6.3973,
|
|
"mean_token_accuracy": 0.134781850874424,
|
|
"num_tokens": 2275462.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.512863445281982,
|
|
"epoch": 1.0825096691018479,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004993411702409948,
|
|
"loss": 6.3671,
|
|
"mean_token_accuracy": 0.136747994273901,
|
|
"num_tokens": 2283826.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.527937221527099,
|
|
"epoch": 1.0868070477009024,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004993155002934002,
|
|
"loss": 6.3742,
|
|
"mean_token_accuracy": 0.13640071898698808,
|
|
"num_tokens": 2292967.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.6068323135375975,
|
|
"epoch": 1.091104426299957,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.0004992893405304111,
|
|
"loss": 6.5223,
|
|
"mean_token_accuracy": 0.13157014027237893,
|
|
"num_tokens": 2302336.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.5013409614562985,
|
|
"epoch": 1.0954018048990115,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00049926269100915,
|
|
"loss": 6.4184,
|
|
"mean_token_accuracy": 0.13976007997989653,
|
|
"num_tokens": 2311465.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.582226037979126,
|
|
"epoch": 1.0996991834980663,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004992355517878087,
|
|
"loss": 6.5322,
|
|
"mean_token_accuracy": 0.13058929219841958,
|
|
"num_tokens": 2320281.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.528811502456665,
|
|
"epoch": 1.1039965620971208,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004992079229256484,
|
|
"loss": 6.4264,
|
|
"mean_token_accuracy": 0.12976702973246573,
|
|
"num_tokens": 2329755.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.4910472393035885,
|
|
"epoch": 1.1082939406961754,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004991798044829996,
|
|
"loss": 6.3767,
|
|
"mean_token_accuracy": 0.13432964831590652,
|
|
"num_tokens": 2338807.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.610663795471192,
|
|
"epoch": 1.11259131929523,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004991511965212618,
|
|
"loss": 6.4672,
|
|
"mean_token_accuracy": 0.13277963325381278,
|
|
"num_tokens": 2348056.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.493175792694092,
|
|
"epoch": 1.1168886978942845,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004991220991029032,
|
|
"loss": 6.4859,
|
|
"mean_token_accuracy": 0.13005012422800064,
|
|
"num_tokens": 2357780.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.564814758300781,
|
|
"epoch": 1.121186076493339,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000499092512291461,
|
|
"loss": 6.4499,
|
|
"mean_token_accuracy": 0.13376415893435478,
|
|
"num_tokens": 2367060.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.5433674335479735,
|
|
"epoch": 1.1254834550923936,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000499062436151541,
|
|
"loss": 6.4314,
|
|
"mean_token_accuracy": 0.13260492384433747,
|
|
"num_tokens": 2375751.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.533551168441773,
|
|
"epoch": 1.129780833691448,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004990318707488173,
|
|
"loss": 6.5099,
|
|
"mean_token_accuracy": 0.1266904965043068,
|
|
"num_tokens": 2385013.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.557472038269043,
|
|
"epoch": 1.1340782122905029,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004990008161500327,
|
|
"loss": 6.3801,
|
|
"mean_token_accuracy": 0.14031620547175408,
|
|
"num_tokens": 2392935.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.503352212905884,
|
|
"epoch": 1.1383755908895574,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000498969272422998,
|
|
"loss": 6.4049,
|
|
"mean_token_accuracy": 0.14203781262040138,
|
|
"num_tokens": 2401560.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.473355054855347,
|
|
"epoch": 1.142672969488612,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004989372396365921,
|
|
"loss": 6.3262,
|
|
"mean_token_accuracy": 0.13685949817299842,
|
|
"num_tokens": 2410050.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.538544416427612,
|
|
"epoch": 1.1469703480876665,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004989047178607618,
|
|
"loss": 6.4325,
|
|
"mean_token_accuracy": 0.136402727663517,
|
|
"num_tokens": 2418980.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.51959867477417,
|
|
"epoch": 1.151267726686721,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004988717071665215,
|
|
"loss": 6.4253,
|
|
"mean_token_accuracy": 0.1328268252313137,
|
|
"num_tokens": 2427992.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.489779472351074,
|
|
"epoch": 1.1555651052857756,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004988382076259537,
|
|
"loss": 6.3409,
|
|
"mean_token_accuracy": 0.13519960269331932,
|
|
"num_tokens": 2436368.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.419038581848144,
|
|
"epoch": 1.1598624838848304,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004988042193122077,
|
|
"loss": 6.3336,
|
|
"mean_token_accuracy": 0.14399233236908912,
|
|
"num_tokens": 2445499.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.545976400375366,
|
|
"epoch": 1.164159862483885,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004987697422995005,
|
|
"loss": 6.3652,
|
|
"mean_token_accuracy": 0.13806355968117714,
|
|
"num_tokens": 2454312.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.509993267059326,
|
|
"epoch": 1.1684572410829395,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004987347766631161,
|
|
"loss": 6.4259,
|
|
"mean_token_accuracy": 0.13582523986697198,
|
|
"num_tokens": 2462922.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.490610218048095,
|
|
"epoch": 1.172754619681994,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004986993224794055,
|
|
"loss": 6.4833,
|
|
"mean_token_accuracy": 0.12911048233509065,
|
|
"num_tokens": 2472195.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.525020790100098,
|
|
"epoch": 1.1770519982810486,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004986633798257865,
|
|
"loss": 6.3819,
|
|
"mean_token_accuracy": 0.13833871781826018,
|
|
"num_tokens": 2481021.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.4724901676177975,
|
|
"epoch": 1.181349376880103,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004986269487807434,
|
|
"loss": 6.3939,
|
|
"mean_token_accuracy": 0.13369214460253714,
|
|
"num_tokens": 2490250.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.521751642227173,
|
|
"epoch": 1.1856467554791577,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000498590029423827,
|
|
"loss": 6.4626,
|
|
"mean_token_accuracy": 0.1330279603600502,
|
|
"num_tokens": 2499122.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.55017466545105,
|
|
"epoch": 1.1899441340782122,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004985526218356546,
|
|
"loss": 6.4269,
|
|
"mean_token_accuracy": 0.13100680485367774,
|
|
"num_tokens": 2508454.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.4670604228973385,
|
|
"epoch": 1.1942415126772667,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004985147260979093,
|
|
"loss": 6.3665,
|
|
"mean_token_accuracy": 0.1456628531217575,
|
|
"num_tokens": 2517353.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.5169178485870365,
|
|
"epoch": 1.1985388912763215,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004984763422933402,
|
|
"loss": 6.3758,
|
|
"mean_token_accuracy": 0.14275188893079757,
|
|
"num_tokens": 2526321.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.4620006561279295,
|
|
"epoch": 1.202836269875376,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004984374705057623,
|
|
"loss": 6.4184,
|
|
"mean_token_accuracy": 0.13711269348859786,
|
|
"num_tokens": 2535924.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.438448047637939,
|
|
"epoch": 1.2071336484744306,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004983981108200561,
|
|
"loss": 6.4106,
|
|
"mean_token_accuracy": 0.13143872320652009,
|
|
"num_tokens": 2545606.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.424742078781128,
|
|
"epoch": 1.2114310270734852,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004983582633221672,
|
|
"loss": 6.3839,
|
|
"mean_token_accuracy": 0.13689693436026573,
|
|
"num_tokens": 2554947.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.620230293273925,
|
|
"epoch": 1.2157284056725397,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004983179280991068,
|
|
"loss": 6.5392,
|
|
"mean_token_accuracy": 0.12959686368703843,
|
|
"num_tokens": 2564462.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.524547338485718,
|
|
"epoch": 1.2200257842715942,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004982771052389508,
|
|
"loss": 6.3797,
|
|
"mean_token_accuracy": 0.141092449426651,
|
|
"num_tokens": 2573124.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.482010555267334,
|
|
"epoch": 1.224323162870649,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004982357948308401,
|
|
"loss": 6.4728,
|
|
"mean_token_accuracy": 0.12631098851561545,
|
|
"num_tokens": 2581829.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.566068410873413,
|
|
"epoch": 1.2286205414697036,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004981939969649799,
|
|
"loss": 6.3285,
|
|
"mean_token_accuracy": 0.13993503898382187,
|
|
"num_tokens": 2590631.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.452121257781982,
|
|
"epoch": 1.232917920068758,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004981517117326404,
|
|
"loss": 6.4557,
|
|
"mean_token_accuracy": 0.13728617280721664,
|
|
"num_tokens": 2600684.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.486893844604492,
|
|
"epoch": 1.2372152986678127,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004981089392261553,
|
|
"loss": 6.3508,
|
|
"mean_token_accuracy": 0.14346109181642533,
|
|
"num_tokens": 2609667.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.4833098411560055,
|
|
"epoch": 1.2415126772668672,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000498065679538923,
|
|
"loss": 6.447,
|
|
"mean_token_accuracy": 0.13537913784384728,
|
|
"num_tokens": 2620025.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.500622749328613,
|
|
"epoch": 1.2458100558659218,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004980219327654049,
|
|
"loss": 6.3454,
|
|
"mean_token_accuracy": 0.13805984035134317,
|
|
"num_tokens": 2629032.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.423587512969971,
|
|
"epoch": 1.2501074344649763,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000497977699001127,
|
|
"loss": 6.3492,
|
|
"mean_token_accuracy": 0.13877770975232123,
|
|
"num_tokens": 2638303.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.512550354003906,
|
|
"epoch": 1.2544048130640308,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004979329783426778,
|
|
"loss": 6.3561,
|
|
"mean_token_accuracy": 0.1412653423845768,
|
|
"num_tokens": 2647902.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.393660068511963,
|
|
"epoch": 1.2587021916630854,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004978877708877094,
|
|
"loss": 6.4141,
|
|
"mean_token_accuracy": 0.13558436930179596,
|
|
"num_tokens": 2657902.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.485628700256347,
|
|
"epoch": 1.2629995702621402,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004978420767349368,
|
|
"loss": 6.3533,
|
|
"mean_token_accuracy": 0.13661525994539261,
|
|
"num_tokens": 2667082.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.461724853515625,
|
|
"epoch": 1.2672969488611947,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004977958959841379,
|
|
"loss": 6.4176,
|
|
"mean_token_accuracy": 0.1369339980185032,
|
|
"num_tokens": 2676855.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.50676383972168,
|
|
"epoch": 1.2715943274602493,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000497749228736153,
|
|
"loss": 6.3514,
|
|
"mean_token_accuracy": 0.13876376375555993,
|
|
"num_tokens": 2685750.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.416869592666626,
|
|
"epoch": 1.2758917060593038,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004977020750928845,
|
|
"loss": 6.433,
|
|
"mean_token_accuracy": 0.1434354990720749,
|
|
"num_tokens": 2695272.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.51352972984314,
|
|
"epoch": 1.2801890846583583,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004976544351572973,
|
|
"loss": 6.3404,
|
|
"mean_token_accuracy": 0.14171317517757415,
|
|
"num_tokens": 2704806.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.346848726272583,
|
|
"epoch": 1.2844864632574131,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004976063090334179,
|
|
"loss": 6.4255,
|
|
"mean_token_accuracy": 0.13577363267540932,
|
|
"num_tokens": 2713521.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.52107515335083,
|
|
"epoch": 1.2887838418564677,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004975576968263346,
|
|
"loss": 6.3883,
|
|
"mean_token_accuracy": 0.13231359645724297,
|
|
"num_tokens": 2721848.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.2887838418564677,
|
|
"eval_entropy": 6.345255539223954,
|
|
"eval_loss": 6.523230075836182,
|
|
"eval_mean_token_accuracy": 0.13721702877130057,
|
|
"eval_num_tokens": 2721848.0,
|
|
"eval_runtime": 2.6548,
|
|
"eval_samples_per_second": 1336.804,
|
|
"eval_steps_per_second": 167.242,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.387936162948608,
|
|
"epoch": 1.2930812204555222,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000497508598642197,
|
|
"loss": 6.3864,
|
|
"mean_token_accuracy": 0.13755422458052635,
|
|
"num_tokens": 2731473.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.494973611831665,
|
|
"epoch": 1.2973785990545768,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000497459014588216,
|
|
"loss": 6.4542,
|
|
"mean_token_accuracy": 0.13196125030517578,
|
|
"num_tokens": 2739867.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.497220182418824,
|
|
"epoch": 1.3016759776536313,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000497408944772663,
|
|
"loss": 6.3679,
|
|
"mean_token_accuracy": 0.13822355195879937,
|
|
"num_tokens": 2748903.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.3612871170043945,
|
|
"epoch": 1.3059733562526858,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004973583893048707,
|
|
"loss": 6.3532,
|
|
"mean_token_accuracy": 0.1409588225185871,
|
|
"num_tokens": 2757711.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.536658239364624,
|
|
"epoch": 1.3102707348517404,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004973073482952321,
|
|
"loss": 6.3729,
|
|
"mean_token_accuracy": 0.13777902573347092,
|
|
"num_tokens": 2765633.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.325454330444336,
|
|
"epoch": 1.314568113450795,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0004972558218552004,
|
|
"loss": 6.3885,
|
|
"mean_token_accuracy": 0.13688459247350693,
|
|
"num_tokens": 2774495.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.457279062271118,
|
|
"epoch": 1.3188654920498495,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004972038100972885,
|
|
"loss": 6.4144,
|
|
"mean_token_accuracy": 0.1356251411139965,
|
|
"num_tokens": 2782665.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.404415273666382,
|
|
"epoch": 1.323162870648904,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004971513131350697,
|
|
"loss": 6.3567,
|
|
"mean_token_accuracy": 0.13246357813477516,
|
|
"num_tokens": 2791394.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.45956335067749,
|
|
"epoch": 1.3274602492479588,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004970983310831759,
|
|
"loss": 6.3453,
|
|
"mean_token_accuracy": 0.13404711484909057,
|
|
"num_tokens": 2800488.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.423970699310303,
|
|
"epoch": 1.3317576278470133,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004970448640572989,
|
|
"loss": 6.4699,
|
|
"mean_token_accuracy": 0.13074692264199256,
|
|
"num_tokens": 2810116.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.48825306892395,
|
|
"epoch": 1.336055006446068,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004969909121741895,
|
|
"loss": 6.2652,
|
|
"mean_token_accuracy": 0.1469486728310585,
|
|
"num_tokens": 2819205.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.351226568222046,
|
|
"epoch": 1.3403523850451224,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004969364755516569,
|
|
"loss": 6.3695,
|
|
"mean_token_accuracy": 0.13450068160891532,
|
|
"num_tokens": 2828017.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.484574985504151,
|
|
"epoch": 1.344649763644177,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004968815543085689,
|
|
"loss": 6.3818,
|
|
"mean_token_accuracy": 0.14407142549753188,
|
|
"num_tokens": 2837125.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.4092048645019535,
|
|
"epoch": 1.3489471422432318,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004968261485648516,
|
|
"loss": 6.4181,
|
|
"mean_token_accuracy": 0.13618151545524598,
|
|
"num_tokens": 2845438.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.4626288414001465,
|
|
"epoch": 1.3532445208422863,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000496770258441489,
|
|
"loss": 6.3725,
|
|
"mean_token_accuracy": 0.14898090660572053,
|
|
"num_tokens": 2854389.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.470159339904785,
|
|
"epoch": 1.3575418994413408,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004967138840605228,
|
|
"loss": 6.3543,
|
|
"mean_token_accuracy": 0.14097132980823518,
|
|
"num_tokens": 2863654.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.292443513870239,
|
|
"epoch": 1.3618392780403954,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000496657025545052,
|
|
"loss": 6.2276,
|
|
"mean_token_accuracy": 0.14617823660373688,
|
|
"num_tokens": 2872871.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.395755338668823,
|
|
"epoch": 1.36613665663945,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000496599683019233,
|
|
"loss": 6.3365,
|
|
"mean_token_accuracy": 0.13821565210819245,
|
|
"num_tokens": 2881140.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.442350387573242,
|
|
"epoch": 1.3704340352385045,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000496541856608279,
|
|
"loss": 6.3427,
|
|
"mean_token_accuracy": 0.13636346384882927,
|
|
"num_tokens": 2889945.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.320608520507813,
|
|
"epoch": 1.374731413837559,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004964835464384595,
|
|
"loss": 6.2689,
|
|
"mean_token_accuracy": 0.14311176240444184,
|
|
"num_tokens": 2898897.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.39306378364563,
|
|
"epoch": 1.3790287924366136,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000496424752637101,
|
|
"loss": 6.2941,
|
|
"mean_token_accuracy": 0.1499859094619751,
|
|
"num_tokens": 2907717.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.33044958114624,
|
|
"epoch": 1.3833261710356681,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004963654753325853,
|
|
"loss": 6.279,
|
|
"mean_token_accuracy": 0.14709102511405944,
|
|
"num_tokens": 2916213.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.484282922744751,
|
|
"epoch": 1.387623549634723,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004963057146543505,
|
|
"loss": 6.4515,
|
|
"mean_token_accuracy": 0.1318092055618763,
|
|
"num_tokens": 2925706.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.370199489593506,
|
|
"epoch": 1.3919209282337774,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00049624547073289,
|
|
"loss": 6.3646,
|
|
"mean_token_accuracy": 0.13615709096193312,
|
|
"num_tokens": 2934464.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.424905824661255,
|
|
"epoch": 1.396218306832832,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004961847436997526,
|
|
"loss": 6.2487,
|
|
"mean_token_accuracy": 0.14671390429139136,
|
|
"num_tokens": 2944095.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.252603054046631,
|
|
"epoch": 1.4005156854318865,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004961235336875416,
|
|
"loss": 6.2806,
|
|
"mean_token_accuracy": 0.148430734872818,
|
|
"num_tokens": 2953357.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.447627115249634,
|
|
"epoch": 1.404813064030941,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004960618408299154,
|
|
"loss": 6.4168,
|
|
"mean_token_accuracy": 0.13321906998753547,
|
|
"num_tokens": 2963020.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.40946683883667,
|
|
"epoch": 1.4091104426299956,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004959996652615865,
|
|
"loss": 6.2535,
|
|
"mean_token_accuracy": 0.1424227312207222,
|
|
"num_tokens": 2971955.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.423822689056396,
|
|
"epoch": 1.4134078212290504,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004959370071183216,
|
|
"loss": 6.3105,
|
|
"mean_token_accuracy": 0.14094186946749687,
|
|
"num_tokens": 2980662.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.450599622726441,
|
|
"epoch": 1.417705199828105,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004958738665369407,
|
|
"loss": 6.468,
|
|
"mean_token_accuracy": 0.12201680019497871,
|
|
"num_tokens": 2990038.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.363819646835327,
|
|
"epoch": 1.4220025784271595,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004958102436553179,
|
|
"loss": 6.3468,
|
|
"mean_token_accuracy": 0.13677727654576302,
|
|
"num_tokens": 2999835.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.500001335144043,
|
|
"epoch": 1.426299957026214,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00049574613861238,
|
|
"loss": 6.3559,
|
|
"mean_token_accuracy": 0.13545409068465233,
|
|
"num_tokens": 3009593.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.450218868255615,
|
|
"epoch": 1.4305973356252686,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004956815515481069,
|
|
"loss": 6.3892,
|
|
"mean_token_accuracy": 0.13879076838493348,
|
|
"num_tokens": 3019187.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.340459680557251,
|
|
"epoch": 1.4348947142243231,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004956164826035309,
|
|
"loss": 6.3029,
|
|
"mean_token_accuracy": 0.139052102714777,
|
|
"num_tokens": 3027875.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.371453809738159,
|
|
"epoch": 1.4391920928233777,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004955509319207363,
|
|
"loss": 6.3297,
|
|
"mean_token_accuracy": 0.14043114185333253,
|
|
"num_tokens": 3036902.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.305567932128906,
|
|
"epoch": 1.4434894714224322,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004954848996428601,
|
|
"loss": 6.3169,
|
|
"mean_token_accuracy": 0.1494467869400978,
|
|
"num_tokens": 3046653.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.4449866771697994,
|
|
"epoch": 1.4477868500214868,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00049541838591409,
|
|
"loss": 6.4144,
|
|
"mean_token_accuracy": 0.13647531121969222,
|
|
"num_tokens": 3056273.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.368440723419189,
|
|
"epoch": 1.4520842286205415,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004953513908796657,
|
|
"loss": 6.2903,
|
|
"mean_token_accuracy": 0.14086930006742476,
|
|
"num_tokens": 3065662.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.460172462463379,
|
|
"epoch": 1.456381607219596,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004952839146858773,
|
|
"loss": 6.3601,
|
|
"mean_token_accuracy": 0.14138378724455833,
|
|
"num_tokens": 3073970.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.337978792190552,
|
|
"epoch": 1.4606789858186506,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004952159574800658,
|
|
"loss": 6.3465,
|
|
"mean_token_accuracy": 0.14081282541155815,
|
|
"num_tokens": 3082500.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.464159679412842,
|
|
"epoch": 1.4649763644177052,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004951475194106229,
|
|
"loss": 6.309,
|
|
"mean_token_accuracy": 0.14396740794181823,
|
|
"num_tokens": 3091574.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.303813648223877,
|
|
"epoch": 1.4692737430167597,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004950786006269898,
|
|
"loss": 6.4083,
|
|
"mean_token_accuracy": 0.13154604732990266,
|
|
"num_tokens": 3102402.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.429269886016845,
|
|
"epoch": 1.4735711216158143,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004950092012796576,
|
|
"loss": 6.2335,
|
|
"mean_token_accuracy": 0.14663561135530473,
|
|
"num_tokens": 3111347.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.318646812438965,
|
|
"epoch": 1.477868500214869,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004949393215201666,
|
|
"loss": 6.2914,
|
|
"mean_token_accuracy": 0.141550412774086,
|
|
"num_tokens": 3120018.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.430832958221435,
|
|
"epoch": 1.4821658788139236,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004948689615011065,
|
|
"loss": 6.3679,
|
|
"mean_token_accuracy": 0.13474266082048417,
|
|
"num_tokens": 3129669.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.336291790008545,
|
|
"epoch": 1.4864632574129781,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004947981213761154,
|
|
"loss": 6.293,
|
|
"mean_token_accuracy": 0.14378871768712997,
|
|
"num_tokens": 3139112.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.382043790817261,
|
|
"epoch": 1.4907606360120327,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004947268012998797,
|
|
"loss": 6.2583,
|
|
"mean_token_accuracy": 0.149325230717659,
|
|
"num_tokens": 3148437.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.344302654266357,
|
|
"epoch": 1.4950580146110872,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000494655001428134,
|
|
"loss": 6.2319,
|
|
"mean_token_accuracy": 0.14777325242757797,
|
|
"num_tokens": 3158165.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.35403413772583,
|
|
"epoch": 1.4993553932101418,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004945827219176604,
|
|
"loss": 6.3396,
|
|
"mean_token_accuracy": 0.14397911876440048,
|
|
"num_tokens": 3167262.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.300922012329101,
|
|
"epoch": 1.5036527718091963,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004945099629262888,
|
|
"loss": 6.3153,
|
|
"mean_token_accuracy": 0.14401653483510019,
|
|
"num_tokens": 3176696.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.516821098327637,
|
|
"epoch": 1.5079501504082509,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004944367246128954,
|
|
"loss": 6.4051,
|
|
"mean_token_accuracy": 0.13158292397856713,
|
|
"num_tokens": 3185857.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.322407150268555,
|
|
"epoch": 1.5122475290073054,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004943630071374036,
|
|
"loss": 6.238,
|
|
"mean_token_accuracy": 0.14783694297075273,
|
|
"num_tokens": 3194687.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.360917568206787,
|
|
"epoch": 1.51654490760636,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004942888106607828,
|
|
"loss": 6.3042,
|
|
"mean_token_accuracy": 0.13809553682804107,
|
|
"num_tokens": 3204913.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.407247304916382,
|
|
"epoch": 1.5208422862054147,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004942141353450486,
|
|
"loss": 6.2806,
|
|
"mean_token_accuracy": 0.1467963233590126,
|
|
"num_tokens": 3213312.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.266504383087158,
|
|
"epoch": 1.5251396648044693,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004941389813532619,
|
|
"loss": 6.1966,
|
|
"mean_token_accuracy": 0.15321253687143327,
|
|
"num_tokens": 3222992.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.292226409912109,
|
|
"epoch": 1.5294370434035238,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000494063348849529,
|
|
"loss": 6.2465,
|
|
"mean_token_accuracy": 0.14908600896596907,
|
|
"num_tokens": 3232836.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.369072008132934,
|
|
"epoch": 1.5337344220025786,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004939872379990011,
|
|
"loss": 6.3969,
|
|
"mean_token_accuracy": 0.1363193489611149,
|
|
"num_tokens": 3243171.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.453199338912964,
|
|
"epoch": 1.5380318006016331,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004939106489678739,
|
|
"loss": 6.3202,
|
|
"mean_token_accuracy": 0.14574317038059234,
|
|
"num_tokens": 3251995.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.290447521209717,
|
|
"epoch": 1.5423291792006877,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000493833581923387,
|
|
"loss": 6.2767,
|
|
"mean_token_accuracy": 0.14543591439723969,
|
|
"num_tokens": 3260841.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.400963926315308,
|
|
"epoch": 1.5466265577997422,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004937560370338244,
|
|
"loss": 6.4211,
|
|
"mean_token_accuracy": 0.12829189896583557,
|
|
"num_tokens": 3270979.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.429856061935425,
|
|
"epoch": 1.5509239363987968,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000493678014468513,
|
|
"loss": 6.3445,
|
|
"mean_token_accuracy": 0.14674683660268784,
|
|
"num_tokens": 3279848.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.351033353805542,
|
|
"epoch": 1.5552213149978513,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004935995143978227,
|
|
"loss": 6.3473,
|
|
"mean_token_accuracy": 0.1431448981165886,
|
|
"num_tokens": 3289172.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.279811143875122,
|
|
"epoch": 1.5595186935969059,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004935205369931664,
|
|
"loss": 6.2238,
|
|
"mean_token_accuracy": 0.14499909579753875,
|
|
"num_tokens": 3297432.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.445178556442261,
|
|
"epoch": 1.5638160721959604,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004934410824269992,
|
|
"loss": 6.2737,
|
|
"mean_token_accuracy": 0.141563268750906,
|
|
"num_tokens": 3307486.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.307324457168579,
|
|
"epoch": 1.568113450795015,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004933611508728182,
|
|
"loss": 6.2366,
|
|
"mean_token_accuracy": 0.1490813337266445,
|
|
"num_tokens": 3316296.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.337137031555176,
|
|
"epoch": 1.5724108293940695,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000493280742505162,
|
|
"loss": 6.2683,
|
|
"mean_token_accuracy": 0.13996259570121766,
|
|
"num_tokens": 3326080.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.304437828063965,
|
|
"epoch": 1.576708207993124,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004931998574996102,
|
|
"loss": 6.1942,
|
|
"mean_token_accuracy": 0.14732773900032042,
|
|
"num_tokens": 3334826.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.278463125228882,
|
|
"epoch": 1.5810055865921788,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004931184960327832,
|
|
"loss": 6.1964,
|
|
"mean_token_accuracy": 0.15309718549251555,
|
|
"num_tokens": 3343261.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.246944999694824,
|
|
"epoch": 1.5853029651912334,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 0.0004930366582823421,
|
|
"loss": 6.254,
|
|
"mean_token_accuracy": 0.13932514265179635,
|
|
"num_tokens": 3352513.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.297404479980469,
|
|
"epoch": 1.589600343790288,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004929543444269879,
|
|
"loss": 6.2721,
|
|
"mean_token_accuracy": 0.1496797688305378,
|
|
"num_tokens": 3361577.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.356855630874634,
|
|
"epoch": 1.5938977223893425,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000492871554646461,
|
|
"loss": 6.3814,
|
|
"mean_token_accuracy": 0.13586916849017144,
|
|
"num_tokens": 3370591.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.315164089202881,
|
|
"epoch": 1.5981951009883972,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004927882891215413,
|
|
"loss": 6.2772,
|
|
"mean_token_accuracy": 0.14198606386780738,
|
|
"num_tokens": 3379761.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.386864757537841,
|
|
"epoch": 1.6024924795874518,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004927045480340475,
|
|
"loss": 6.3681,
|
|
"mean_token_accuracy": 0.13762572556734085,
|
|
"num_tokens": 3388974.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.3296960353851315,
|
|
"epoch": 1.6067898581865063,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004926203315668363,
|
|
"loss": 6.2626,
|
|
"mean_token_accuracy": 0.14389025494456292,
|
|
"num_tokens": 3398339.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.360977649688721,
|
|
"epoch": 1.6110872367855609,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004925356399038032,
|
|
"loss": 6.2436,
|
|
"mean_token_accuracy": 0.14258339628577232,
|
|
"num_tokens": 3408292.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.300831747055054,
|
|
"epoch": 1.6153846153846154,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004924504732298808,
|
|
"loss": 6.2229,
|
|
"mean_token_accuracy": 0.14572942852973939,
|
|
"num_tokens": 3417057.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.337365961074829,
|
|
"epoch": 1.61968199398367,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004923648317310391,
|
|
"loss": 6.3284,
|
|
"mean_token_accuracy": 0.13823174461722373,
|
|
"num_tokens": 3425830.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.3136725425720215,
|
|
"epoch": 1.6239793725827245,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004922787155942849,
|
|
"loss": 6.3579,
|
|
"mean_token_accuracy": 0.13356941863894461,
|
|
"num_tokens": 3435513.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.36700325012207,
|
|
"epoch": 1.628276751181779,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004921921250076611,
|
|
"loss": 6.263,
|
|
"mean_token_accuracy": 0.1448775663971901,
|
|
"num_tokens": 3444684.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.274904537200928,
|
|
"epoch": 1.6325741297808336,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004921050601602475,
|
|
"loss": 6.3363,
|
|
"mean_token_accuracy": 0.1468764767050743,
|
|
"num_tokens": 3453454.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.317870998382569,
|
|
"epoch": 1.6368715083798882,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004920175212421587,
|
|
"loss": 6.2725,
|
|
"mean_token_accuracy": 0.14472019597887992,
|
|
"num_tokens": 3463228.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.330316781997681,
|
|
"epoch": 1.6411688869789427,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004919295084445445,
|
|
"loss": 6.1527,
|
|
"mean_token_accuracy": 0.14830412566661835,
|
|
"num_tokens": 3472131.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.2725340843200685,
|
|
"epoch": 1.6454662655779975,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004918410219595899,
|
|
"loss": 6.216,
|
|
"mean_token_accuracy": 0.14919435530900954,
|
|
"num_tokens": 3480642.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.415084791183472,
|
|
"epoch": 1.649763644177052,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000491752061980514,
|
|
"loss": 6.2119,
|
|
"mean_token_accuracy": 0.1491432771086693,
|
|
"num_tokens": 3489346.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.201951742172241,
|
|
"epoch": 1.6540610227761066,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004916626287015697,
|
|
"loss": 6.2543,
|
|
"mean_token_accuracy": 0.14897455722093583,
|
|
"num_tokens": 3498473.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.358420991897583,
|
|
"epoch": 1.658358401375161,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004915727223180436,
|
|
"loss": 6.2574,
|
|
"mean_token_accuracy": 0.14309950396418572,
|
|
"num_tokens": 3507415.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.299440002441406,
|
|
"epoch": 1.6626557799742159,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004914823430262554,
|
|
"loss": 6.3807,
|
|
"mean_token_accuracy": 0.13274637162685393,
|
|
"num_tokens": 3516873.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.339557313919068,
|
|
"epoch": 1.6669531585732704,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004913914910235573,
|
|
"loss": 6.2218,
|
|
"mean_token_accuracy": 0.15021957010030745,
|
|
"num_tokens": 3525047.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.251499080657959,
|
|
"epoch": 1.671250537172325,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004913001665083337,
|
|
"loss": 6.2479,
|
|
"mean_token_accuracy": 0.14710038602352143,
|
|
"num_tokens": 3534354.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.303897285461426,
|
|
"epoch": 1.6755479157713795,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004912083696800008,
|
|
"loss": 6.2738,
|
|
"mean_token_accuracy": 0.13878671005368232,
|
|
"num_tokens": 3543830.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.299686336517334,
|
|
"epoch": 1.679845294370434,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004911161007390063,
|
|
"loss": 6.1846,
|
|
"mean_token_accuracy": 0.13919671922922133,
|
|
"num_tokens": 3552314.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.231910467147827,
|
|
"epoch": 1.6841426729694886,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004910233598868287,
|
|
"loss": 6.2644,
|
|
"mean_token_accuracy": 0.14159709364175796,
|
|
"num_tokens": 3561656.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.353840303421021,
|
|
"epoch": 1.6884400515685432,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004909301473259769,
|
|
"loss": 6.2569,
|
|
"mean_token_accuracy": 0.1448962077498436,
|
|
"num_tokens": 3571784.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.361128616333008,
|
|
"epoch": 1.6927374301675977,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004908364632599899,
|
|
"loss": 6.2222,
|
|
"mean_token_accuracy": 0.14719251841306685,
|
|
"num_tokens": 3580626.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.236987209320068,
|
|
"epoch": 1.6970348087666522,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004907423078934362,
|
|
"loss": 6.2312,
|
|
"mean_token_accuracy": 0.14682269841432571,
|
|
"num_tokens": 3589916.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.305332803726197,
|
|
"epoch": 1.7013321873657068,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004906476814319134,
|
|
"loss": 6.2369,
|
|
"mean_token_accuracy": 0.14632309079170228,
|
|
"num_tokens": 3599128.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.226185417175293,
|
|
"epoch": 1.7056295659647613,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004905525840820481,
|
|
"loss": 6.2551,
|
|
"mean_token_accuracy": 0.14322940409183502,
|
|
"num_tokens": 3608764.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.428410005569458,
|
|
"epoch": 1.709926944563816,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004904570160514948,
|
|
"loss": 6.3085,
|
|
"mean_token_accuracy": 0.13539518415927887,
|
|
"num_tokens": 3619082.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.2587316036224365,
|
|
"epoch": 1.7142243231628707,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004903609775489358,
|
|
"loss": 6.2504,
|
|
"mean_token_accuracy": 0.14486311376094818,
|
|
"num_tokens": 3628695.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.297107553482055,
|
|
"epoch": 1.7185217017619252,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004902644687840809,
|
|
"loss": 6.2379,
|
|
"mean_token_accuracy": 0.1429143115878105,
|
|
"num_tokens": 3637599.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.7185217017619252,
|
|
"eval_entropy": 6.115066128808099,
|
|
"eval_loss": 6.353027820587158,
|
|
"eval_mean_token_accuracy": 0.14586634908777638,
|
|
"eval_num_tokens": 3637599.0,
|
|
"eval_runtime": 2.662,
|
|
"eval_samples_per_second": 1333.19,
|
|
"eval_steps_per_second": 166.79,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.2849133014678955,
|
|
"epoch": 1.7228190803609797,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004901674899676667,
|
|
"loss": 6.2368,
|
|
"mean_token_accuracy": 0.1443016067147255,
|
|
"num_tokens": 3647406.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.2112464427948,
|
|
"epoch": 1.7271164589600345,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004900700413114561,
|
|
"loss": 6.1097,
|
|
"mean_token_accuracy": 0.14899515807628633,
|
|
"num_tokens": 3656531.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.229878091812134,
|
|
"epoch": 1.731413837559089,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000489972123028238,
|
|
"loss": 6.2296,
|
|
"mean_token_accuracy": 0.14038500413298607,
|
|
"num_tokens": 3664922.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.331856203079224,
|
|
"epoch": 1.7357112161581436,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004898737353318268,
|
|
"loss": 6.1483,
|
|
"mean_token_accuracy": 0.1493585541844368,
|
|
"num_tokens": 3673283.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.2273845195770265,
|
|
"epoch": 1.7400085947571982,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000489774878437062,
|
|
"loss": 6.3012,
|
|
"mean_token_accuracy": 0.14301757216453553,
|
|
"num_tokens": 3681760.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.2626869678497314,
|
|
"epoch": 1.7443059733562527,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004896755525598074,
|
|
"loss": 6.1037,
|
|
"mean_token_accuracy": 0.14743346273899077,
|
|
"num_tokens": 3689408.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.244910812377929,
|
|
"epoch": 1.7486033519553073,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004895757579169511,
|
|
"loss": 6.2301,
|
|
"mean_token_accuracy": 0.1434326946735382,
|
|
"num_tokens": 3697904.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.350824546813965,
|
|
"epoch": 1.7529007305543618,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004894754947264047,
|
|
"loss": 6.2576,
|
|
"mean_token_accuracy": 0.1488267719745636,
|
|
"num_tokens": 3706704.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.315699911117553,
|
|
"epoch": 1.7571981091534163,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000489374763207103,
|
|
"loss": 6.3224,
|
|
"mean_token_accuracy": 0.13804764896631241,
|
|
"num_tokens": 3715690.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.324631690979004,
|
|
"epoch": 1.761495487752471,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004892735635790033,
|
|
"loss": 6.1191,
|
|
"mean_token_accuracy": 0.1548789396882057,
|
|
"num_tokens": 3724835.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.214821243286133,
|
|
"epoch": 1.7657928663515254,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000489171896063085,
|
|
"loss": 6.1576,
|
|
"mean_token_accuracy": 0.15082096308469772,
|
|
"num_tokens": 3733977.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.289961481094361,
|
|
"epoch": 1.77009024495058,
|
|
"grad_norm": 2.796875,
|
|
"learning_rate": 0.0004890697608813495,
|
|
"loss": 6.2704,
|
|
"mean_token_accuracy": 0.13823474571108818,
|
|
"num_tokens": 3742665.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.396904516220093,
|
|
"epoch": 1.7743876235496348,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004889671582568193,
|
|
"loss": 6.3399,
|
|
"mean_token_accuracy": 0.14141005724668504,
|
|
"num_tokens": 3751647.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.215439128875732,
|
|
"epoch": 1.7786850021486893,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004888640884135374,
|
|
"loss": 6.2358,
|
|
"mean_token_accuracy": 0.14285386800765992,
|
|
"num_tokens": 3760852.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.276702117919922,
|
|
"epoch": 1.7829823807477438,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004887605515765671,
|
|
"loss": 6.1812,
|
|
"mean_token_accuracy": 0.14567631036043166,
|
|
"num_tokens": 3768640.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.216250801086426,
|
|
"epoch": 1.7872797593467986,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004886565479719914,
|
|
"loss": 6.2124,
|
|
"mean_token_accuracy": 0.14191268533468246,
|
|
"num_tokens": 3776504.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.39114670753479,
|
|
"epoch": 1.7915771379458532,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004885520778269128,
|
|
"loss": 6.26,
|
|
"mean_token_accuracy": 0.14373670518398285,
|
|
"num_tokens": 3786353.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.258846092224121,
|
|
"epoch": 1.7958745165449077,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004884471413694523,
|
|
"loss": 6.2893,
|
|
"mean_token_accuracy": 0.14175898283720018,
|
|
"num_tokens": 3795902.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.270934009552002,
|
|
"epoch": 1.8001718951439623,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004883417388287491,
|
|
"loss": 6.2008,
|
|
"mean_token_accuracy": 0.14041119515895845,
|
|
"num_tokens": 3805986.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.298763370513916,
|
|
"epoch": 1.8044692737430168,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004882358704349603,
|
|
"loss": 6.3162,
|
|
"mean_token_accuracy": 0.14934238642454148,
|
|
"num_tokens": 3814915.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.270471668243408,
|
|
"epoch": 1.8087666523420713,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004881295364192601,
|
|
"loss": 6.2026,
|
|
"mean_token_accuracy": 0.14982720762491225,
|
|
"num_tokens": 3823966.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.377622318267822,
|
|
"epoch": 1.813064030941126,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004880227370138394,
|
|
"loss": 6.2571,
|
|
"mean_token_accuracy": 0.14232114627957343,
|
|
"num_tokens": 3832775.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.227845954895019,
|
|
"epoch": 1.8173614095401804,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004879154724519057,
|
|
"loss": 6.1892,
|
|
"mean_token_accuracy": 0.14971126914024352,
|
|
"num_tokens": 3842808.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.318926382064819,
|
|
"epoch": 1.821658788139235,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004878077429676816,
|
|
"loss": 6.3262,
|
|
"mean_token_accuracy": 0.14237651824951172,
|
|
"num_tokens": 3853303.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.271210956573486,
|
|
"epoch": 1.8259561667382895,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004876995487964054,
|
|
"loss": 6.2375,
|
|
"mean_token_accuracy": 0.13341685086488725,
|
|
"num_tokens": 3862462.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.280560922622681,
|
|
"epoch": 1.830253545337344,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00048759089017432996,
|
|
"loss": 6.3441,
|
|
"mean_token_accuracy": 0.13644540086388587,
|
|
"num_tokens": 3871596.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.344761610031128,
|
|
"epoch": 1.8345509239363988,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004874817673387222,
|
|
"loss": 6.2636,
|
|
"mean_token_accuracy": 0.1410408765077591,
|
|
"num_tokens": 3881276.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.186103487014771,
|
|
"epoch": 1.8388483025354534,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048737218052786275,
|
|
"loss": 6.3125,
|
|
"mean_token_accuracy": 0.1408373661339283,
|
|
"num_tokens": 3891610.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.385894393920898,
|
|
"epoch": 1.843145681134508,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048726212998104554,
|
|
"loss": 6.2444,
|
|
"mean_token_accuracy": 0.13924626559019088,
|
|
"num_tokens": 3900584.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.247769498825074,
|
|
"epoch": 1.8474430597335625,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004871516159385768,
|
|
"loss": 6.1962,
|
|
"mean_token_accuracy": 0.14932045117020606,
|
|
"num_tokens": 3910208.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.08746452331543,
|
|
"epoch": 1.8517404383326173,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004870406386417752,
|
|
"loss": 6.1231,
|
|
"mean_token_accuracy": 0.15434293150901796,
|
|
"num_tokens": 3918424.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.275387907028199,
|
|
"epoch": 1.8560378169316718,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004869291983329707,
|
|
"loss": 6.0612,
|
|
"mean_token_accuracy": 0.15940360426902772,
|
|
"num_tokens": 3926206.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.293136215209961,
|
|
"epoch": 1.8603351955307263,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004868172952555044,
|
|
"loss": 6.1597,
|
|
"mean_token_accuracy": 0.14148508310317992,
|
|
"num_tokens": 3935769.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.198759841918945,
|
|
"epoch": 1.864632574129781,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004867049296537278,
|
|
"loss": 6.1383,
|
|
"mean_token_accuracy": 0.14691901803016663,
|
|
"num_tokens": 3945118.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.259021234512329,
|
|
"epoch": 1.8689299527288354,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004865921017730027,
|
|
"loss": 6.2437,
|
|
"mean_token_accuracy": 0.14798983931541443,
|
|
"num_tokens": 3954012.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.239273023605347,
|
|
"epoch": 1.87322733132789,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048647881185969995,
|
|
"loss": 6.257,
|
|
"mean_token_accuracy": 0.1421762339770794,
|
|
"num_tokens": 3964239.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.246995878219605,
|
|
"epoch": 1.8775247099269445,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004863650601611994,
|
|
"loss": 6.1484,
|
|
"mean_token_accuracy": 0.14947519302368165,
|
|
"num_tokens": 3973694.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.242728757858276,
|
|
"epoch": 1.881822088525999,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00048625084692588937,
|
|
"loss": 6.2014,
|
|
"mean_token_accuracy": 0.15101856291294097,
|
|
"num_tokens": 3982706.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.219791555404663,
|
|
"epoch": 1.8861194671250536,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00048613617240316593,
|
|
"loss": 6.1553,
|
|
"mean_token_accuracy": 0.1522357404232025,
|
|
"num_tokens": 3990934.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.30114893913269,
|
|
"epoch": 1.8904168457241082,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004860210368434323,
|
|
"loss": 6.1962,
|
|
"mean_token_accuracy": 0.15202590972185134,
|
|
"num_tokens": 3999864.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.231817770004272,
|
|
"epoch": 1.8947142243231627,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048590544049809857,
|
|
"loss": 6.1912,
|
|
"mean_token_accuracy": 0.15230269283056258,
|
|
"num_tokens": 4008273.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.250846576690674,
|
|
"epoch": 1.8990116029222175,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000485789383619581,
|
|
"loss": 6.2226,
|
|
"mean_token_accuracy": 0.14662364721298218,
|
|
"num_tokens": 4017697.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.260433864593506,
|
|
"epoch": 1.903308981521272,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004856728664613015,
|
|
"loss": 6.24,
|
|
"mean_token_accuracy": 0.14280799701809882,
|
|
"num_tokens": 4026775.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.199958467483521,
|
|
"epoch": 1.9076063601203266,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00048555588927768674,
|
|
"loss": 6.2271,
|
|
"mean_token_accuracy": 0.14681547209620477,
|
|
"num_tokens": 4036476.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.2801882266998295,
|
|
"epoch": 1.9119037387193811,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004854384523241683,
|
|
"loss": 6.1932,
|
|
"mean_token_accuracy": 0.1446580171585083,
|
|
"num_tokens": 4045221.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.186880826950073,
|
|
"epoch": 1.916201117318436,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00048532055585718143,
|
|
"loss": 6.099,
|
|
"mean_token_accuracy": 0.15187990218400954,
|
|
"num_tokens": 4053754.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.19896125793457,
|
|
"epoch": 1.9204984959174904,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00048520220013416505,
|
|
"loss": 6.1748,
|
|
"mean_token_accuracy": 0.14861172139644624,
|
|
"num_tokens": 4061730.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.253337478637695,
|
|
"epoch": 1.924795874516545,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004850833854135607,
|
|
"loss": 6.2257,
|
|
"mean_token_accuracy": 0.1466619372367859,
|
|
"num_tokens": 4070501.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.280824756622314,
|
|
"epoch": 1.9290932531155995,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004849641119548122,
|
|
"loss": 6.3067,
|
|
"mean_token_accuracy": 0.14085772708058358,
|
|
"num_tokens": 4079621.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.368407249450684,
|
|
"epoch": 1.933390631714654,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000484844380018365,
|
|
"loss": 6.3004,
|
|
"mean_token_accuracy": 0.14154468029737471,
|
|
"num_tokens": 4090106.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.3014167785644535,
|
|
"epoch": 1.9376880103137086,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000484724189865666,
|
|
"loss": 6.2223,
|
|
"mean_token_accuracy": 0.14424855262041092,
|
|
"num_tokens": 4099269.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.0989758491516115,
|
|
"epoch": 1.9419853889127632,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004846035417591624,
|
|
"loss": 6.1476,
|
|
"mean_token_accuracy": 0.15749251544475557,
|
|
"num_tokens": 4108414.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.340393972396851,
|
|
"epoch": 1.9462827675118177,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004844824359623014,
|
|
"loss": 6.2867,
|
|
"mean_token_accuracy": 0.13618101105093955,
|
|
"num_tokens": 4117731.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.304265356063842,
|
|
"epoch": 1.9505801461108723,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00048436087273952966,
|
|
"loss": 6.2463,
|
|
"mean_token_accuracy": 0.14030979871749877,
|
|
"num_tokens": 4127194.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.237520790100097,
|
|
"epoch": 1.9548775247099268,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00048423885235629265,
|
|
"loss": 6.2093,
|
|
"mean_token_accuracy": 0.15297809690237046,
|
|
"num_tokens": 4135594.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.249992609024048,
|
|
"epoch": 1.9591749033089814,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004841163750790342,
|
|
"loss": 6.2415,
|
|
"mean_token_accuracy": 0.14738751128315924,
|
|
"num_tokens": 4145027.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.196361923217774,
|
|
"epoch": 1.9634722819080361,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00048399344117519555,
|
|
"loss": 6.1207,
|
|
"mean_token_accuracy": 0.15311533063650132,
|
|
"num_tokens": 4153754.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.201253414154053,
|
|
"epoch": 1.9677696605070907,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00048387005091321544,
|
|
"loss": 6.1878,
|
|
"mean_token_accuracy": 0.150342957675457,
|
|
"num_tokens": 4162765.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.319298219680786,
|
|
"epoch": 1.9720670391061452,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00048374620456252877,
|
|
"loss": 6.199,
|
|
"mean_token_accuracy": 0.14727609902620314,
|
|
"num_tokens": 4171589.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.2408675193786625,
|
|
"epoch": 1.9763644177052,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00048362190239356644,
|
|
"loss": 6.2104,
|
|
"mean_token_accuracy": 0.14729856103658676,
|
|
"num_tokens": 4181817.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.207037305831909,
|
|
"epoch": 1.9806617963042545,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048349714467775474,
|
|
"loss": 6.1466,
|
|
"mean_token_accuracy": 0.14971105605363846,
|
|
"num_tokens": 4191350.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.247203063964844,
|
|
"epoch": 1.984959174903309,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00048337193168751464,
|
|
"loss": 6.2228,
|
|
"mean_token_accuracy": 0.14333939254283906,
|
|
"num_tokens": 4199888.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.260077619552613,
|
|
"epoch": 1.9892565535023636,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004832462636962613,
|
|
"loss": 6.1929,
|
|
"mean_token_accuracy": 0.14196112975478173,
|
|
"num_tokens": 4209509.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.231012678146362,
|
|
"epoch": 1.9935539321014182,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004831201409784034,
|
|
"loss": 6.1546,
|
|
"mean_token_accuracy": 0.14981688112020491,
|
|
"num_tokens": 4218496.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.250877714157104,
|
|
"epoch": 1.9978513107004727,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004829935638093424,
|
|
"loss": 6.1503,
|
|
"mean_token_accuracy": 0.1517081841826439,
|
|
"num_tokens": 4227504.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.2071972952948675,
|
|
"epoch": 2.0017189514396216,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004828665324654724,
|
|
"loss": 6.0675,
|
|
"mean_token_accuracy": 0.1486931476328108,
|
|
"num_tokens": 4235338.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.284207487106324,
|
|
"epoch": 2.006016330038676,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004827390472241791,
|
|
"loss": 5.8336,
|
|
"mean_token_accuracy": 0.15509926825761794,
|
|
"num_tokens": 4244905.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.201587200164795,
|
|
"epoch": 2.010313708637731,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004826111083638392,
|
|
"loss": 5.933,
|
|
"mean_token_accuracy": 0.15552402809262275,
|
|
"num_tokens": 4254533.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.175057792663575,
|
|
"epoch": 2.0146110872367857,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00048248271616382,
|
|
"loss": 5.8866,
|
|
"mean_token_accuracy": 0.15550699681043625,
|
|
"num_tokens": 4264023.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.124167585372925,
|
|
"epoch": 2.0189084658358403,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00048235387090447894,
|
|
"loss": 5.9287,
|
|
"mean_token_accuracy": 0.15373642817139627,
|
|
"num_tokens": 4273298.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.234945964813233,
|
|
"epoch": 2.023205844434895,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048222457286716235,
|
|
"loss": 5.8777,
|
|
"mean_token_accuracy": 0.1560370445251465,
|
|
"num_tokens": 4283244.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.134788990020752,
|
|
"epoch": 2.0275032230339494,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00048209482233420564,
|
|
"loss": 5.8203,
|
|
"mean_token_accuracy": 0.16626692563295364,
|
|
"num_tokens": 4291677.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.193808126449585,
|
|
"epoch": 2.031800601633004,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000481964619588932,
|
|
"loss": 5.8819,
|
|
"mean_token_accuracy": 0.15690354257822037,
|
|
"num_tokens": 4300822.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.134625625610352,
|
|
"epoch": 2.0360979802320585,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004818339649156523,
|
|
"loss": 5.9028,
|
|
"mean_token_accuracy": 0.15883450508117675,
|
|
"num_tokens": 4310149.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.0754838466644285,
|
|
"epoch": 2.040395358831113,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00048170285859966395,
|
|
"loss": 5.8058,
|
|
"mean_token_accuracy": 0.1661190629005432,
|
|
"num_tokens": 4319109.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.145545196533203,
|
|
"epoch": 2.0446927374301676,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048157130092725087,
|
|
"loss": 5.7847,
|
|
"mean_token_accuracy": 0.16278679817914962,
|
|
"num_tokens": 4327921.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.176446056365966,
|
|
"epoch": 2.048990116029222,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004814392921856824,
|
|
"loss": 5.9416,
|
|
"mean_token_accuracy": 0.1598782941699028,
|
|
"num_tokens": 4338026.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.094702672958374,
|
|
"epoch": 2.0532874946282766,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004813068326632128,
|
|
"loss": 5.7988,
|
|
"mean_token_accuracy": 0.16492851078510284,
|
|
"num_tokens": 4347794.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.16729211807251,
|
|
"epoch": 2.057584873227331,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004811739226490809,
|
|
"loss": 5.9722,
|
|
"mean_token_accuracy": 0.1584267795085907,
|
|
"num_tokens": 4357249.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.173635578155517,
|
|
"epoch": 2.0618822518263857,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00048104056243350896,
|
|
"loss": 5.9148,
|
|
"mean_token_accuracy": 0.15438505113124848,
|
|
"num_tokens": 4366053.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.123129224777221,
|
|
"epoch": 2.0661796304254403,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004809067523077023,
|
|
"loss": 5.9267,
|
|
"mean_token_accuracy": 0.15736443251371385,
|
|
"num_tokens": 4375543.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.103467178344727,
|
|
"epoch": 2.0704770090244953,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00048077249256384884,
|
|
"loss": 5.8027,
|
|
"mean_token_accuracy": 0.16944397389888763,
|
|
"num_tokens": 4384332.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.0881589412689205,
|
|
"epoch": 2.07477438762355,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004806377834951182,
|
|
"loss": 5.8964,
|
|
"mean_token_accuracy": 0.15095557272434235,
|
|
"num_tokens": 4393670.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.219787645339966,
|
|
"epoch": 2.0790717662226044,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00048050262539566104,
|
|
"loss": 5.8947,
|
|
"mean_token_accuracy": 0.16174632906913758,
|
|
"num_tokens": 4402763.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.077099752426148,
|
|
"epoch": 2.083369144821659,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004803670185606087,
|
|
"loss": 5.8197,
|
|
"mean_token_accuracy": 0.1643015444278717,
|
|
"num_tokens": 4411863.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.1350139617919925,
|
|
"epoch": 2.0876665234207135,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004802309632860724,
|
|
"loss": 5.901,
|
|
"mean_token_accuracy": 0.15897406488656998,
|
|
"num_tokens": 4421110.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.1598176002502445,
|
|
"epoch": 2.091963902019768,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00048009445986914236,
|
|
"loss": 5.9105,
|
|
"mean_token_accuracy": 0.15521485656499862,
|
|
"num_tokens": 4430249.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.089999341964722,
|
|
"epoch": 2.0962612806188226,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047995750860788756,
|
|
"loss": 5.8766,
|
|
"mean_token_accuracy": 0.15184871554374696,
|
|
"num_tokens": 4439686.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.125515270233154,
|
|
"epoch": 2.100558659217877,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004798201098013547,
|
|
"loss": 5.854,
|
|
"mean_token_accuracy": 0.16153764277696608,
|
|
"num_tokens": 4448645.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.100013065338135,
|
|
"epoch": 2.1048560378169316,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047968226374956797,
|
|
"loss": 5.827,
|
|
"mean_token_accuracy": 0.1589351251721382,
|
|
"num_tokens": 4456870.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.031601905822754,
|
|
"epoch": 2.109153416415986,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00047954397075352794,
|
|
"loss": 5.8683,
|
|
"mean_token_accuracy": 0.16552610844373702,
|
|
"num_tokens": 4466287.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.109052085876465,
|
|
"epoch": 2.1134507950150407,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00047940523111521136,
|
|
"loss": 5.7589,
|
|
"mean_token_accuracy": 0.1646905466914177,
|
|
"num_tokens": 4474461.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.135332489013672,
|
|
"epoch": 2.1177481736140953,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0004792660451375701,
|
|
"loss": 5.8604,
|
|
"mean_token_accuracy": 0.15848599076271058,
|
|
"num_tokens": 4483002.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.047499132156372,
|
|
"epoch": 2.12204555221315,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00047912641312453064,
|
|
"loss": 5.8544,
|
|
"mean_token_accuracy": 0.15858358442783355,
|
|
"num_tokens": 4492061.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.1697187423706055,
|
|
"epoch": 2.1263429308122044,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047898633538099363,
|
|
"loss": 5.9077,
|
|
"mean_token_accuracy": 0.15018413960933685,
|
|
"num_tokens": 4501829.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.152231645584107,
|
|
"epoch": 2.130640309411259,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004788458122128327,
|
|
"loss": 5.9283,
|
|
"mean_token_accuracy": 0.15399457588791848,
|
|
"num_tokens": 4511539.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.088359832763672,
|
|
"epoch": 2.134937688010314,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047870484392689434,
|
|
"loss": 5.7892,
|
|
"mean_token_accuracy": 0.16311607509851456,
|
|
"num_tokens": 4520425.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.094126749038696,
|
|
"epoch": 2.1392350666093685,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000478563430830997,
|
|
"loss": 5.8859,
|
|
"mean_token_accuracy": 0.15439863502979279,
|
|
"num_tokens": 4529474.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.147933292388916,
|
|
"epoch": 2.143532445208423,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047842157323393035,
|
|
"loss": 5.8346,
|
|
"mean_token_accuracy": 0.1593649595975876,
|
|
"num_tokens": 4538082.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.073186922073364,
|
|
"epoch": 2.1478298238074776,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004782792714454547,
|
|
"loss": 6.0273,
|
|
"mean_token_accuracy": 0.15312504917383193,
|
|
"num_tokens": 4547340.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.1478298238074776,
|
|
"eval_entropy": 6.05996425302179,
|
|
"eval_loss": 6.275111675262451,
|
|
"eval_mean_token_accuracy": 0.1505782189904838,
|
|
"eval_num_tokens": 4547340.0,
|
|
"eval_runtime": 2.6554,
|
|
"eval_samples_per_second": 1336.521,
|
|
"eval_steps_per_second": 167.206,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 6.184024429321289,
|
|
"epoch": 2.152127202406532,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004781365257763002,
|
|
"loss": 5.8816,
|
|
"mean_token_accuracy": 0.15708422362804414,
|
|
"num_tokens": 4556415.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.035333919525146,
|
|
"epoch": 2.1564245810055866,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00047799333653816633,
|
|
"loss": 5.7422,
|
|
"mean_token_accuracy": 0.16808970272541046,
|
|
"num_tokens": 4565156.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.091459465026856,
|
|
"epoch": 2.160721959604641,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00047784970404372124,
|
|
"loss": 5.8418,
|
|
"mean_token_accuracy": 0.16465607434511184,
|
|
"num_tokens": 4574678.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.0676025390625,
|
|
"epoch": 2.1650193382036957,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00047770562860660083,
|
|
"loss": 5.8768,
|
|
"mean_token_accuracy": 0.15344698950648308,
|
|
"num_tokens": 4583253.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.100003910064697,
|
|
"epoch": 2.1693167168027503,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004775611105414083,
|
|
"loss": 5.9405,
|
|
"mean_token_accuracy": 0.15367247760295868,
|
|
"num_tokens": 4594042.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.073186540603638,
|
|
"epoch": 2.173614095401805,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004774161501637133,
|
|
"loss": 5.8833,
|
|
"mean_token_accuracy": 0.1599620595574379,
|
|
"num_tokens": 4603128.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.064374351501465,
|
|
"epoch": 2.1779114740008594,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004772707477900514,
|
|
"loss": 5.8725,
|
|
"mean_token_accuracy": 0.1599646419286728,
|
|
"num_tokens": 4611537.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 6.202516984939575,
|
|
"epoch": 2.182208852599914,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004771249037379232,
|
|
"loss": 5.988,
|
|
"mean_token_accuracy": 0.15234462320804595,
|
|
"num_tokens": 4622481.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.065233516693115,
|
|
"epoch": 2.1865062311989685,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004769786183257939,
|
|
"loss": 5.885,
|
|
"mean_token_accuracy": 0.16063564866781235,
|
|
"num_tokens": 4631259.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.159190988540649,
|
|
"epoch": 2.190803609798023,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004768318918730924,
|
|
"loss": 5.8271,
|
|
"mean_token_accuracy": 0.16417029201984407,
|
|
"num_tokens": 4640266.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.971075344085693,
|
|
"epoch": 2.195100988397078,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00047668472470021044,
|
|
"loss": 5.9024,
|
|
"mean_token_accuracy": 0.15217453688383104,
|
|
"num_tokens": 4649520.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 6.112492609024048,
|
|
"epoch": 2.1993983669961326,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004765371171285025,
|
|
"loss": 5.8177,
|
|
"mean_token_accuracy": 0.16666681468486785,
|
|
"num_tokens": 4658501.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.071633672714233,
|
|
"epoch": 2.203695745595187,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00047638906948028445,
|
|
"loss": 5.8806,
|
|
"mean_token_accuracy": 0.1598675012588501,
|
|
"num_tokens": 4667567.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 6.059203004837036,
|
|
"epoch": 2.2079931241942417,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00047624058207883317,
|
|
"loss": 5.8755,
|
|
"mean_token_accuracy": 0.16072613298892974,
|
|
"num_tokens": 4676618.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 6.196649360656738,
|
|
"epoch": 2.212290502793296,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00047609165524838576,
|
|
"loss": 5.9503,
|
|
"mean_token_accuracy": 0.1545178085565567,
|
|
"num_tokens": 4685967.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 6.03065505027771,
|
|
"epoch": 2.2165878813923507,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.0004759422893141389,
|
|
"loss": 5.8446,
|
|
"mean_token_accuracy": 0.15830490738153458,
|
|
"num_tokens": 4694568.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.054008817672729,
|
|
"epoch": 2.2208852599914053,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004757924846022482,
|
|
"loss": 5.906,
|
|
"mean_token_accuracy": 0.15464629977941513,
|
|
"num_tokens": 4703648.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.095048713684082,
|
|
"epoch": 2.22518263859046,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00047564224143982714,
|
|
"loss": 5.7653,
|
|
"mean_token_accuracy": 0.16790542900562286,
|
|
"num_tokens": 4712444.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.145687198638916,
|
|
"epoch": 2.2294800171895144,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00047549156015494676,
|
|
"loss": 5.9307,
|
|
"mean_token_accuracy": 0.1549680009484291,
|
|
"num_tokens": 4722034.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 6.0166431903839115,
|
|
"epoch": 2.233777395788569,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 0.00047534044107663484,
|
|
"loss": 5.9212,
|
|
"mean_token_accuracy": 0.15793955326080322,
|
|
"num_tokens": 4731344.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.1388434886932375,
|
|
"epoch": 2.2380747743876235,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00047518888453487496,
|
|
"loss": 5.835,
|
|
"mean_token_accuracy": 0.16322563588619232,
|
|
"num_tokens": 4739302.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 6.104530334472656,
|
|
"epoch": 2.242372152986678,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.0004750368908606061,
|
|
"loss": 5.9644,
|
|
"mean_token_accuracy": 0.15467115491628647,
|
|
"num_tokens": 4748848.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 6.097955894470215,
|
|
"epoch": 2.2466695315857326,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00047488446038572164,
|
|
"loss": 6.0145,
|
|
"mean_token_accuracy": 0.1517575427889824,
|
|
"num_tokens": 4758194.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.208459997177124,
|
|
"epoch": 2.250966910184787,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0004747315934430688,
|
|
"loss": 5.9274,
|
|
"mean_token_accuracy": 0.15489244610071182,
|
|
"num_tokens": 4768081.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 6.0705572128295895,
|
|
"epoch": 2.2552642887838417,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000474578290366448,
|
|
"loss": 5.8479,
|
|
"mean_token_accuracy": 0.15939725041389466,
|
|
"num_tokens": 4776471.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.072135639190674,
|
|
"epoch": 2.259561667382896,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004744245514906117,
|
|
"loss": 5.8774,
|
|
"mean_token_accuracy": 0.16055589616298677,
|
|
"num_tokens": 4784403.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.9916871070861815,
|
|
"epoch": 2.263859045981951,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00047427037715126426,
|
|
"loss": 5.8252,
|
|
"mean_token_accuracy": 0.1523001715540886,
|
|
"num_tokens": 4792779.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.997291326522827,
|
|
"epoch": 2.2681564245810057,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004741157676850608,
|
|
"loss": 5.8294,
|
|
"mean_token_accuracy": 0.1603853225708008,
|
|
"num_tokens": 4801426.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 6.146497631072998,
|
|
"epoch": 2.2724538031800603,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 0.00047396072342960663,
|
|
"loss": 5.8712,
|
|
"mean_token_accuracy": 0.15908808559179305,
|
|
"num_tokens": 4810329.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.057448244094848,
|
|
"epoch": 2.276751181779115,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00047380524472345645,
|
|
"loss": 5.9283,
|
|
"mean_token_accuracy": 0.15035422146320343,
|
|
"num_tokens": 4819544.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.065110683441162,
|
|
"epoch": 2.2810485603781694,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004736493319061134,
|
|
"loss": 5.9307,
|
|
"mean_token_accuracy": 0.15485063940286636,
|
|
"num_tokens": 4828113.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 6.082950162887573,
|
|
"epoch": 2.285345938977224,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004734929853180291,
|
|
"loss": 5.8973,
|
|
"mean_token_accuracy": 0.15679322630167009,
|
|
"num_tokens": 4836989.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.119369077682495,
|
|
"epoch": 2.2896433175762785,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047333620530060175,
|
|
"loss": 5.9513,
|
|
"mean_token_accuracy": 0.1539230540394783,
|
|
"num_tokens": 4847103.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 6.102674579620361,
|
|
"epoch": 2.293940696175333,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004731789921961764,
|
|
"loss": 5.9603,
|
|
"mean_token_accuracy": 0.15662157386541367,
|
|
"num_tokens": 4856238.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.13006181716919,
|
|
"epoch": 2.2982380747743876,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004730213463480434,
|
|
"loss": 5.8604,
|
|
"mean_token_accuracy": 0.16454241126775743,
|
|
"num_tokens": 4864608.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.052781105041504,
|
|
"epoch": 2.302535453373442,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00047286326810043857,
|
|
"loss": 5.8065,
|
|
"mean_token_accuracy": 0.16103093922138215,
|
|
"num_tokens": 4873889.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.985052442550659,
|
|
"epoch": 2.3068328319724967,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00047270475779854137,
|
|
"loss": 5.872,
|
|
"mean_token_accuracy": 0.16748757511377335,
|
|
"num_tokens": 4882902.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.193141222000122,
|
|
"epoch": 2.311130210571551,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00047254581578847507,
|
|
"loss": 5.8907,
|
|
"mean_token_accuracy": 0.15794829428195953,
|
|
"num_tokens": 4892390.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 6.04741907119751,
|
|
"epoch": 2.3154275891706058,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004723864424173055,
|
|
"loss": 6.0077,
|
|
"mean_token_accuracy": 0.15141431987285614,
|
|
"num_tokens": 4901625.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.090276861190796,
|
|
"epoch": 2.3197249677696608,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004722266380330403,
|
|
"loss": 5.8036,
|
|
"mean_token_accuracy": 0.1645648568868637,
|
|
"num_tokens": 4910804.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 6.069483709335327,
|
|
"epoch": 2.3240223463687153,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047206640298462857,
|
|
"loss": 5.8718,
|
|
"mean_token_accuracy": 0.16099148318171502,
|
|
"num_tokens": 4920441.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.066246461868286,
|
|
"epoch": 2.32831972496777,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00047190573762195945,
|
|
"loss": 5.9129,
|
|
"mean_token_accuracy": 0.15659259110689164,
|
|
"num_tokens": 4930204.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.068677473068237,
|
|
"epoch": 2.3326171035668244,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00047174464229586186,
|
|
"loss": 6.0247,
|
|
"mean_token_accuracy": 0.15192122906446456,
|
|
"num_tokens": 4941191.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 6.157104206085205,
|
|
"epoch": 2.336914482165879,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0004715831173581036,
|
|
"loss": 5.9944,
|
|
"mean_token_accuracy": 0.15300290137529374,
|
|
"num_tokens": 4951825.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 6.0795214653015135,
|
|
"epoch": 2.3412118607649335,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00047142116316139073,
|
|
"loss": 5.9231,
|
|
"mean_token_accuracy": 0.15817170441150666,
|
|
"num_tokens": 4960632.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.19065523147583,
|
|
"epoch": 2.345509239363988,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004712587800593663,
|
|
"loss": 5.9458,
|
|
"mean_token_accuracy": 0.15569401681423187,
|
|
"num_tokens": 4969455.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.065016508102417,
|
|
"epoch": 2.3498066179630426,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004710959684066102,
|
|
"loss": 5.8766,
|
|
"mean_token_accuracy": 0.1586523950099945,
|
|
"num_tokens": 4978997.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 6.113391017913818,
|
|
"epoch": 2.354103996562097,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00047093272855863803,
|
|
"loss": 5.916,
|
|
"mean_token_accuracy": 0.15666621774435044,
|
|
"num_tokens": 4988305.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.98998236656189,
|
|
"epoch": 2.3584013751611517,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004707690608719003,
|
|
"loss": 5.8461,
|
|
"mean_token_accuracy": 0.1616642251610756,
|
|
"num_tokens": 4997022.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 6.12568883895874,
|
|
"epoch": 2.362698753760206,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004706049657037818,
|
|
"loss": 5.9231,
|
|
"mean_token_accuracy": 0.15089340060949324,
|
|
"num_tokens": 5005664.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 6.077770185470581,
|
|
"epoch": 2.3669961323592608,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004704404434126009,
|
|
"loss": 5.8754,
|
|
"mean_token_accuracy": 0.15403562039136887,
|
|
"num_tokens": 5014769.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 6.0987646102905275,
|
|
"epoch": 2.3712935109583153,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047027549435760843,
|
|
"loss": 5.9403,
|
|
"mean_token_accuracy": 0.15741077959537506,
|
|
"num_tokens": 5024060.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.18181586265564,
|
|
"epoch": 2.37559088955737,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0004701101188989872,
|
|
"loss": 5.9872,
|
|
"mean_token_accuracy": 0.1542619563639164,
|
|
"num_tokens": 5033046.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.0726195812225345,
|
|
"epoch": 2.3798882681564244,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00046994431739785114,
|
|
"loss": 5.8285,
|
|
"mean_token_accuracy": 0.16705856174230577,
|
|
"num_tokens": 5040894.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.050960302352905,
|
|
"epoch": 2.384185646755479,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00046977809021624454,
|
|
"loss": 6.0064,
|
|
"mean_token_accuracy": 0.15674740076065063,
|
|
"num_tokens": 5050961.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.1043839931488035,
|
|
"epoch": 2.3884830253545335,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004696114377171409,
|
|
"loss": 5.9261,
|
|
"mean_token_accuracy": 0.15404321253299713,
|
|
"num_tokens": 5060226.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 6.0166253566741945,
|
|
"epoch": 2.3927804039535885,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004694443602644429,
|
|
"loss": 5.8801,
|
|
"mean_token_accuracy": 0.1530889317393303,
|
|
"num_tokens": 5069225.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.132807922363281,
|
|
"epoch": 2.397077782552643,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004692768582229808,
|
|
"loss": 5.8841,
|
|
"mean_token_accuracy": 0.16195421069860458,
|
|
"num_tokens": 5078386.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.026313066482544,
|
|
"epoch": 2.4013751611516976,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00046910893195851213,
|
|
"loss": 5.8087,
|
|
"mean_token_accuracy": 0.15549368262290955,
|
|
"num_tokens": 5087161.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 6.016099739074707,
|
|
"epoch": 2.405672539750752,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00046894058183772074,
|
|
"loss": 5.9579,
|
|
"mean_token_accuracy": 0.1559508666396141,
|
|
"num_tokens": 5096613.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.081924772262573,
|
|
"epoch": 2.4099699183498067,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000468771808228216,
|
|
"loss": 5.9215,
|
|
"mean_token_accuracy": 0.157737098634243,
|
|
"num_tokens": 5106534.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.06602783203125,
|
|
"epoch": 2.414267296948861,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00046860261149853197,
|
|
"loss": 5.9509,
|
|
"mean_token_accuracy": 0.15083148926496506,
|
|
"num_tokens": 5115975.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.993266630172729,
|
|
"epoch": 2.4185646755479158,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004684329920181268,
|
|
"loss": 5.8402,
|
|
"mean_token_accuracy": 0.16399779468774794,
|
|
"num_tokens": 5124635.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.100220823287964,
|
|
"epoch": 2.4228620541469703,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 0.00046826295015738154,
|
|
"loss": 5.8314,
|
|
"mean_token_accuracy": 0.16892157196998597,
|
|
"num_tokens": 5133226.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 6.054720783233643,
|
|
"epoch": 2.427159432746025,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004680924862875996,
|
|
"loss": 5.9125,
|
|
"mean_token_accuracy": 0.15770741254091264,
|
|
"num_tokens": 5142257.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 6.048891830444336,
|
|
"epoch": 2.4314568113450794,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00046792160078100605,
|
|
"loss": 5.8979,
|
|
"mean_token_accuracy": 0.16061281487345697,
|
|
"num_tokens": 5150752.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.037526082992554,
|
|
"epoch": 2.435754189944134,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00046775029401074653,
|
|
"loss": 5.7967,
|
|
"mean_token_accuracy": 0.1623912572860718,
|
|
"num_tokens": 5160237.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.054116344451904,
|
|
"epoch": 2.4400515685431885,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00046757856635088645,
|
|
"loss": 5.894,
|
|
"mean_token_accuracy": 0.16464148461818695,
|
|
"num_tokens": 5169752.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.995990848541259,
|
|
"epoch": 2.444348947142243,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004674064181764105,
|
|
"loss": 5.9163,
|
|
"mean_token_accuracy": 0.16457967907190324,
|
|
"num_tokens": 5178892.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.148447895050049,
|
|
"epoch": 2.448646325741298,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046723384986322147,
|
|
"loss": 5.9012,
|
|
"mean_token_accuracy": 0.15619142353534698,
|
|
"num_tokens": 5188468.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 6.007736349105835,
|
|
"epoch": 2.4529437043403526,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004670608617881395,
|
|
"loss": 5.8192,
|
|
"mean_token_accuracy": 0.16597600281238556,
|
|
"num_tokens": 5197565.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.9673144817352295,
|
|
"epoch": 2.457241082939407,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004668874543289014,
|
|
"loss": 5.8335,
|
|
"mean_token_accuracy": 0.16931346654891968,
|
|
"num_tokens": 5205791.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 6.024132347106933,
|
|
"epoch": 2.4615384615384617,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00046671362786415986,
|
|
"loss": 5.837,
|
|
"mean_token_accuracy": 0.17491899728775023,
|
|
"num_tokens": 5214773.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.978415155410767,
|
|
"epoch": 2.465835840137516,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00046653938277348237,
|
|
"loss": 5.8574,
|
|
"mean_token_accuracy": 0.17243983894586562,
|
|
"num_tokens": 5223734.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.178035831451416,
|
|
"epoch": 2.4701332187365708,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004663647194373505,
|
|
"loss": 5.9335,
|
|
"mean_token_accuracy": 0.1516405776143074,
|
|
"num_tokens": 5231742.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.982483530044556,
|
|
"epoch": 2.4744305973356253,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00046618963823715913,
|
|
"loss": 5.896,
|
|
"mean_token_accuracy": 0.16158574670553208,
|
|
"num_tokens": 5241673.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.007695198059082,
|
|
"epoch": 2.47872797593468,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00046601413955521575,
|
|
"loss": 5.8715,
|
|
"mean_token_accuracy": 0.15949235782027243,
|
|
"num_tokens": 5250082.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.102848672866822,
|
|
"epoch": 2.4830253545337344,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004658382237747393,
|
|
"loss": 5.9239,
|
|
"mean_token_accuracy": 0.16041397750377656,
|
|
"num_tokens": 5259680.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 6.043904638290405,
|
|
"epoch": 2.487322733132789,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046566189127985946,
|
|
"loss": 5.9038,
|
|
"mean_token_accuracy": 0.16174552738666534,
|
|
"num_tokens": 5269561.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.059132623672485,
|
|
"epoch": 2.4916201117318435,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000465485142455616,
|
|
"loss": 5.8758,
|
|
"mean_token_accuracy": 0.16595080643892288,
|
|
"num_tokens": 5278659.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.973031759262085,
|
|
"epoch": 2.495917490330898,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00046530797768795765,
|
|
"loss": 5.8523,
|
|
"mean_token_accuracy": 0.16781924664974213,
|
|
"num_tokens": 5287619.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 6.01035509109497,
|
|
"epoch": 2.5002148689299526,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00046513039736374153,
|
|
"loss": 5.9675,
|
|
"mean_token_accuracy": 0.15401696264743805,
|
|
"num_tokens": 5297334.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 6.1172713279724125,
|
|
"epoch": 2.504512247529007,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004649524018707319,
|
|
"loss": 5.8613,
|
|
"mean_token_accuracy": 0.1616441547870636,
|
|
"num_tokens": 5306208.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.979982185363769,
|
|
"epoch": 2.5088096261280617,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00046477399159759996,
|
|
"loss": 5.7957,
|
|
"mean_token_accuracy": 0.16605262756347655,
|
|
"num_tokens": 5314754.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.916423749923706,
|
|
"epoch": 2.5131070047271162,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00046459516693392246,
|
|
"loss": 5.8404,
|
|
"mean_token_accuracy": 0.16251906752586365,
|
|
"num_tokens": 5324000.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.110569858551026,
|
|
"epoch": 2.517404383326171,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004644159282701808,
|
|
"loss": 5.8964,
|
|
"mean_token_accuracy": 0.15983645617961884,
|
|
"num_tokens": 5332478.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.078077602386474,
|
|
"epoch": 2.5217017619252258,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00046423627599776076,
|
|
"loss": 5.9383,
|
|
"mean_token_accuracy": 0.15348622053861619,
|
|
"num_tokens": 5341635.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 6.057946062088012,
|
|
"epoch": 2.5259991405242803,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000464056210508951,
|
|
"loss": 5.9553,
|
|
"mean_token_accuracy": 0.15539063960313798,
|
|
"num_tokens": 5350144.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 6.064607810974121,
|
|
"epoch": 2.530296519123335,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004638757321969426,
|
|
"loss": 5.868,
|
|
"mean_token_accuracy": 0.16377780735492706,
|
|
"num_tokens": 5358788.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 6.058841848373413,
|
|
"epoch": 2.5345938977223894,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00046369484145582815,
|
|
"loss": 5.9698,
|
|
"mean_token_accuracy": 0.1512514442205429,
|
|
"num_tokens": 5368057.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.996830463409424,
|
|
"epoch": 2.538891276321444,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00046351353868060054,
|
|
"loss": 5.8228,
|
|
"mean_token_accuracy": 0.1620754837989807,
|
|
"num_tokens": 5376739.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 6.0612822532653805,
|
|
"epoch": 2.5431886549204985,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00046333182426715273,
|
|
"loss": 5.9163,
|
|
"mean_token_accuracy": 0.16040000319480896,
|
|
"num_tokens": 5385967.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 6.071706533432007,
|
|
"epoch": 2.547486033519553,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00046314969861227626,
|
|
"loss": 5.9393,
|
|
"mean_token_accuracy": 0.1499171920120716,
|
|
"num_tokens": 5395192.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 6.048627328872681,
|
|
"epoch": 2.5517834121186076,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004629671621136608,
|
|
"loss": 5.9147,
|
|
"mean_token_accuracy": 0.16159721165895463,
|
|
"num_tokens": 5404694.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.040609884262085,
|
|
"epoch": 2.556080790717662,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004627842151698931,
|
|
"loss": 5.9237,
|
|
"mean_token_accuracy": 0.16203372329473495,
|
|
"num_tokens": 5413102.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 6.064021253585816,
|
|
"epoch": 2.5603781693167167,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00046260085818045625,
|
|
"loss": 5.9192,
|
|
"mean_token_accuracy": 0.15512176528573035,
|
|
"num_tokens": 5423339.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 6.064020299911499,
|
|
"epoch": 2.5646755479157712,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004624170915457284,
|
|
"loss": 5.9177,
|
|
"mean_token_accuracy": 0.15929645001888276,
|
|
"num_tokens": 5432377.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 6.003238248825073,
|
|
"epoch": 2.5689729265148262,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00046223291566698264,
|
|
"loss": 5.8294,
|
|
"mean_token_accuracy": 0.16039082184433937,
|
|
"num_tokens": 5441038.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 6.0270661354064945,
|
|
"epoch": 2.5732703051138808,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004620483309463855,
|
|
"loss": 5.8716,
|
|
"mean_token_accuracy": 0.1595895379781723,
|
|
"num_tokens": 5449557.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.081511974334717,
|
|
"epoch": 2.5775676837129353,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004618633377869961,
|
|
"loss": 5.9724,
|
|
"mean_token_accuracy": 0.15905023962259293,
|
|
"num_tokens": 5458931.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.5775676837129353,
|
|
"eval_entropy": 5.9907188479964795,
|
|
"eval_loss": 6.211490154266357,
|
|
"eval_mean_token_accuracy": 0.15491128753165942,
|
|
"eval_num_tokens": 5458931.0,
|
|
"eval_runtime": 2.844,
|
|
"eval_samples_per_second": 1247.869,
|
|
"eval_steps_per_second": 156.116,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 6.051160478591919,
|
|
"epoch": 2.58186506231199,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004616779365927656,
|
|
"loss": 5.812,
|
|
"mean_token_accuracy": 0.1746028408408165,
|
|
"num_tokens": 5468539.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.897491931915283,
|
|
"epoch": 2.5861624409110444,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004614921277685361,
|
|
"loss": 5.7403,
|
|
"mean_token_accuracy": 0.17282533943653106,
|
|
"num_tokens": 5475710.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 6.008006381988525,
|
|
"epoch": 2.590459819510099,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046130591172003976,
|
|
"loss": 5.8894,
|
|
"mean_token_accuracy": 0.15739246606826782,
|
|
"num_tokens": 5484597.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 6.092322254180909,
|
|
"epoch": 2.5947571981091535,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004611192888538981,
|
|
"loss": 5.9551,
|
|
"mean_token_accuracy": 0.15505467355251312,
|
|
"num_tokens": 5493213.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 6.083768463134765,
|
|
"epoch": 2.599054576708208,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00046093225957762084,
|
|
"loss": 5.97,
|
|
"mean_token_accuracy": 0.15912253707647322,
|
|
"num_tokens": 5502556.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 6.054582262039185,
|
|
"epoch": 2.6033519553072626,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004607448242996051,
|
|
"loss": 5.886,
|
|
"mean_token_accuracy": 0.16483262032270432,
|
|
"num_tokens": 5511779.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 6.100365686416626,
|
|
"epoch": 2.607649333906317,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004605569834291347,
|
|
"loss": 5.8748,
|
|
"mean_token_accuracy": 0.16575300991535186,
|
|
"num_tokens": 5520836.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.97064700126648,
|
|
"epoch": 2.6119467125053717,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00046036873737637904,
|
|
"loss": 5.8883,
|
|
"mean_token_accuracy": 0.16223816573619843,
|
|
"num_tokens": 5529285.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 6.031012964248657,
|
|
"epoch": 2.6162440911044262,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004601800865523921,
|
|
"loss": 5.889,
|
|
"mean_token_accuracy": 0.16434804648160933,
|
|
"num_tokens": 5538160.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 6.027869701385498,
|
|
"epoch": 2.620541469703481,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00045999103136911204,
|
|
"loss": 5.8852,
|
|
"mean_token_accuracy": 0.15403481125831603,
|
|
"num_tokens": 5547355.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 6.043389558792114,
|
|
"epoch": 2.6248388483025353,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00045980157223935965,
|
|
"loss": 5.9053,
|
|
"mean_token_accuracy": 0.15608059614896774,
|
|
"num_tokens": 5557299.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.999477767944336,
|
|
"epoch": 2.62913622690159,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00045961170957683806,
|
|
"loss": 5.8302,
|
|
"mean_token_accuracy": 0.16516596525907518,
|
|
"num_tokens": 5565469.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 6.054600429534912,
|
|
"epoch": 2.6334336055006444,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00045942144379613147,
|
|
"loss": 5.9438,
|
|
"mean_token_accuracy": 0.15425471663475038,
|
|
"num_tokens": 5574740.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 6.076825523376465,
|
|
"epoch": 2.637730984099699,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045923077531270426,
|
|
"loss": 5.9107,
|
|
"mean_token_accuracy": 0.16228047907352447,
|
|
"num_tokens": 5583438.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 6.063165760040283,
|
|
"epoch": 2.6420283626987535,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004590397045429001,
|
|
"loss": 5.8951,
|
|
"mean_token_accuracy": 0.16245172768831254,
|
|
"num_tokens": 5592389.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.994216060638427,
|
|
"epoch": 2.646325741297808,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045884823190394134,
|
|
"loss": 5.7902,
|
|
"mean_token_accuracy": 0.17125492542982101,
|
|
"num_tokens": 5601598.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.996226453781128,
|
|
"epoch": 2.650623119896863,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004586563578139275,
|
|
"loss": 5.9033,
|
|
"mean_token_accuracy": 0.15935150533914566,
|
|
"num_tokens": 5610498.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.963082695007325,
|
|
"epoch": 2.6549204984959176,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00045846408269183505,
|
|
"loss": 5.7998,
|
|
"mean_token_accuracy": 0.17070116400718688,
|
|
"num_tokens": 5620082.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 6.059010076522827,
|
|
"epoch": 2.659217877094972,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00045827140695751603,
|
|
"loss": 5.8939,
|
|
"mean_token_accuracy": 0.1581853449344635,
|
|
"num_tokens": 5630291.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 6.0179871082305905,
|
|
"epoch": 2.6635152556940267,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004580783310316971,
|
|
"loss": 5.8703,
|
|
"mean_token_accuracy": 0.16777567714452743,
|
|
"num_tokens": 5638784.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.925207090377808,
|
|
"epoch": 2.6678126342930812,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00045788485533597895,
|
|
"loss": 5.7395,
|
|
"mean_token_accuracy": 0.16571872681379318,
|
|
"num_tokens": 5647968.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 6.08448224067688,
|
|
"epoch": 2.672110012892136,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00045769098029283526,
|
|
"loss": 5.9731,
|
|
"mean_token_accuracy": 0.15026847794651985,
|
|
"num_tokens": 5657543.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 6.030013132095337,
|
|
"epoch": 2.6764073914911903,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004574967063256115,
|
|
"loss": 5.8756,
|
|
"mean_token_accuracy": 0.16706451326608657,
|
|
"num_tokens": 5666535.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 6.058322525024414,
|
|
"epoch": 2.680704770090245,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00045730203385852447,
|
|
"loss": 5.9601,
|
|
"mean_token_accuracy": 0.15754017159342765,
|
|
"num_tokens": 5676273.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.9861531257629395,
|
|
"epoch": 2.6850021486892994,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000457106963316661,
|
|
"loss": 5.8527,
|
|
"mean_token_accuracy": 0.16199901849031448,
|
|
"num_tokens": 5684888.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 6.045160484313965,
|
|
"epoch": 2.689299527288354,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00045691149512597717,
|
|
"loss": 5.9169,
|
|
"mean_token_accuracy": 0.16329616606235503,
|
|
"num_tokens": 5693626.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 6.027765083312988,
|
|
"epoch": 2.6935969058874085,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 0.00045671562971329736,
|
|
"loss": 5.8325,
|
|
"mean_token_accuracy": 0.16559524610638618,
|
|
"num_tokens": 5702542.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.983704090118408,
|
|
"epoch": 2.6978942844864635,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00045651936750631337,
|
|
"loss": 5.8524,
|
|
"mean_token_accuracy": 0.1660203978419304,
|
|
"num_tokens": 5711440.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 6.0599102020263675,
|
|
"epoch": 2.702191663085518,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00045632270893358333,
|
|
"loss": 5.932,
|
|
"mean_token_accuracy": 0.15755877643823624,
|
|
"num_tokens": 5721495.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 6.0640490531921385,
|
|
"epoch": 2.7064890416845726,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004561256544245312,
|
|
"loss": 5.9627,
|
|
"mean_token_accuracy": 0.15352778658270835,
|
|
"num_tokens": 5730664.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 6.013172912597656,
|
|
"epoch": 2.710786420283627,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000455928204409445,
|
|
"loss": 5.8514,
|
|
"mean_token_accuracy": 0.1656482771039009,
|
|
"num_tokens": 5740229.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.992557144165039,
|
|
"epoch": 2.7150837988826817,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00045573035931947684,
|
|
"loss": 5.8284,
|
|
"mean_token_accuracy": 0.1656378298997879,
|
|
"num_tokens": 5748549.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.978826189041138,
|
|
"epoch": 2.7193811774817362,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0004555321195866411,
|
|
"loss": 5.7746,
|
|
"mean_token_accuracy": 0.15985333025455475,
|
|
"num_tokens": 5757603.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 6.037001752853394,
|
|
"epoch": 2.723678556080791,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.0004553334856438143,
|
|
"loss": 5.9606,
|
|
"mean_token_accuracy": 0.15452521443367004,
|
|
"num_tokens": 5767520.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 6.078950643539429,
|
|
"epoch": 2.7279759346798453,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00045513445792473356,
|
|
"loss": 5.9533,
|
|
"mean_token_accuracy": 0.15213411152362824,
|
|
"num_tokens": 5776778.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 6.055939626693726,
|
|
"epoch": 2.7322733132789,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004549350368639958,
|
|
"loss": 6.0085,
|
|
"mean_token_accuracy": 0.15094553530216218,
|
|
"num_tokens": 5785652.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 6.0882477283477785,
|
|
"epoch": 2.7365706918779544,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00045473522289705693,
|
|
"loss": 5.9394,
|
|
"mean_token_accuracy": 0.1568639397621155,
|
|
"num_tokens": 5795766.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.97335991859436,
|
|
"epoch": 2.740868070477009,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00045453501646023085,
|
|
"loss": 5.9785,
|
|
"mean_token_accuracy": 0.15723178684711456,
|
|
"num_tokens": 5804504.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 6.0407976627349855,
|
|
"epoch": 2.7451654490760635,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00045433441799068837,
|
|
"loss": 5.9174,
|
|
"mean_token_accuracy": 0.1563822202384472,
|
|
"num_tokens": 5814161.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.99520959854126,
|
|
"epoch": 2.749462827675118,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004541334279264562,
|
|
"loss": 5.7905,
|
|
"mean_token_accuracy": 0.17239678055047988,
|
|
"num_tokens": 5822235.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.96739559173584,
|
|
"epoch": 2.7537602062741726,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00045393204670641656,
|
|
"loss": 5.8106,
|
|
"mean_token_accuracy": 0.16272899359464646,
|
|
"num_tokens": 5831572.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.955503845214844,
|
|
"epoch": 2.758057584873227,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0004537302747703055,
|
|
"loss": 5.8459,
|
|
"mean_token_accuracy": 0.16618359237909316,
|
|
"num_tokens": 5839694.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 6.13696255683899,
|
|
"epoch": 2.7623549634722817,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00045352811255871216,
|
|
"loss": 5.9509,
|
|
"mean_token_accuracy": 0.16227137297391891,
|
|
"num_tokens": 5849131.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 6.109014368057251,
|
|
"epoch": 2.7666523420713363,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00045332556051307804,
|
|
"loss": 5.8768,
|
|
"mean_token_accuracy": 0.15744196772575378,
|
|
"num_tokens": 5858861.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 6.0058550357818605,
|
|
"epoch": 2.770949720670391,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00045312261907569585,
|
|
"loss": 5.8753,
|
|
"mean_token_accuracy": 0.15888426154851915,
|
|
"num_tokens": 5867585.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 6.001376533508301,
|
|
"epoch": 2.775247099269446,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00045291928868970867,
|
|
"loss": 5.9002,
|
|
"mean_token_accuracy": 0.16079236567020416,
|
|
"num_tokens": 5876256.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.9930966854095455,
|
|
"epoch": 2.7795444778685003,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004527155697991087,
|
|
"loss": 5.966,
|
|
"mean_token_accuracy": 0.1554612174630165,
|
|
"num_tokens": 5885302.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 6.02942476272583,
|
|
"epoch": 2.783841856467555,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004525114628487365,
|
|
"loss": 5.958,
|
|
"mean_token_accuracy": 0.15460113883018495,
|
|
"num_tokens": 5895066.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 6.083170604705811,
|
|
"epoch": 2.7881392350666094,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00045230696828428026,
|
|
"loss": 5.9297,
|
|
"mean_token_accuracy": 0.16147802621126175,
|
|
"num_tokens": 5903258.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 6.028832674026489,
|
|
"epoch": 2.792436613665664,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004521020865522742,
|
|
"loss": 5.8351,
|
|
"mean_token_accuracy": 0.16377991437911987,
|
|
"num_tokens": 5911714.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 6.023666286468506,
|
|
"epoch": 2.7967339922647185,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00045189681810009827,
|
|
"loss": 5.9005,
|
|
"mean_token_accuracy": 0.1569843351840973,
|
|
"num_tokens": 5920432.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 6.088962459564209,
|
|
"epoch": 2.801031370863773,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00045169116337597653,
|
|
"loss": 5.9328,
|
|
"mean_token_accuracy": 0.1598918095231056,
|
|
"num_tokens": 5929202.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 6.084402179718017,
|
|
"epoch": 2.8053287494628276,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.000451485122828977,
|
|
"loss": 5.9791,
|
|
"mean_token_accuracy": 0.15051311552524566,
|
|
"num_tokens": 5938034.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 6.021539163589478,
|
|
"epoch": 2.809626128061882,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00045127869690900956,
|
|
"loss": 5.8118,
|
|
"mean_token_accuracy": 0.16473996192216872,
|
|
"num_tokens": 5946944.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.979176950454712,
|
|
"epoch": 2.8139235066609367,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00045107188606682613,
|
|
"loss": 5.9023,
|
|
"mean_token_accuracy": 0.16275302916765214,
|
|
"num_tokens": 5956475.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 6.0721758842468265,
|
|
"epoch": 2.8182208852599913,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0004508646907540188,
|
|
"loss": 5.8792,
|
|
"mean_token_accuracy": 0.1519957035779953,
|
|
"num_tokens": 5965814.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 6.031784391403198,
|
|
"epoch": 2.8225182638590463,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0004506571114230195,
|
|
"loss": 5.9167,
|
|
"mean_token_accuracy": 0.15551358461380005,
|
|
"num_tokens": 5973850.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.9177404880523685,
|
|
"epoch": 2.826815642458101,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00045044914852709824,
|
|
"loss": 5.8682,
|
|
"mean_token_accuracy": 0.15748999118804932,
|
|
"num_tokens": 5982987.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 6.085504388809204,
|
|
"epoch": 2.8311130210571553,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 0.0004502408025203631,
|
|
"loss": 5.8674,
|
|
"mean_token_accuracy": 0.16563252210617066,
|
|
"num_tokens": 5992227.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.991399621963501,
|
|
"epoch": 2.83541039965621,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004500320738577584,
|
|
"loss": 5.8596,
|
|
"mean_token_accuracy": 0.16823477447032928,
|
|
"num_tokens": 6000243.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.987527751922608,
|
|
"epoch": 2.8397077782552644,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00044982296299506407,
|
|
"loss": 5.8484,
|
|
"mean_token_accuracy": 0.16697341799736024,
|
|
"num_tokens": 6009771.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 6.069508504867554,
|
|
"epoch": 2.844005156854319,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004496134703888948,
|
|
"loss": 5.922,
|
|
"mean_token_accuracy": 0.16114695966243744,
|
|
"num_tokens": 6018683.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.989667463302612,
|
|
"epoch": 2.8483025354533735,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00044940359649669846,
|
|
"loss": 5.7774,
|
|
"mean_token_accuracy": 0.17218757867813111,
|
|
"num_tokens": 6027422.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.9799083232879635,
|
|
"epoch": 2.852599914052428,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00044919334177675595,
|
|
"loss": 5.8689,
|
|
"mean_token_accuracy": 0.16289890110492705,
|
|
"num_tokens": 6035670.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 6.012589025497436,
|
|
"epoch": 2.8568972926514826,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00044898270668817955,
|
|
"loss": 5.7941,
|
|
"mean_token_accuracy": 0.16796652972698212,
|
|
"num_tokens": 6044092.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.979690408706665,
|
|
"epoch": 2.861194671250537,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.000448771691690912,
|
|
"loss": 5.8713,
|
|
"mean_token_accuracy": 0.15729259997606276,
|
|
"num_tokens": 6053970.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 6.033042573928833,
|
|
"epoch": 2.8654920498495917,
|
|
"grad_norm": 1.9296875,
|
|
"learning_rate": 0.0004485602972457257,
|
|
"loss": 5.8297,
|
|
"mean_token_accuracy": 0.16680487841367722,
|
|
"num_tokens": 6062965.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 6.1102265357971195,
|
|
"epoch": 2.8697894284486463,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00044834852381422165,
|
|
"loss": 5.9048,
|
|
"mean_token_accuracy": 0.16165395677089692,
|
|
"num_tokens": 6072420.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.920516061782837,
|
|
"epoch": 2.874086807047701,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00044813637185882836,
|
|
"loss": 5.8081,
|
|
"mean_token_accuracy": 0.1592453435063362,
|
|
"num_tokens": 6080915.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.995471715927124,
|
|
"epoch": 2.8783841856467554,
|
|
"grad_norm": 1.890625,
|
|
"learning_rate": 0.00044792384184280106,
|
|
"loss": 5.9514,
|
|
"mean_token_accuracy": 0.15751733928918837,
|
|
"num_tokens": 6090453.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 6.0667096138000485,
|
|
"epoch": 2.88268156424581,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00044771093423022013,
|
|
"loss": 5.9823,
|
|
"mean_token_accuracy": 0.14896320402622223,
|
|
"num_tokens": 6099390.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.997150564193726,
|
|
"epoch": 2.8869789428448644,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004474976494859909,
|
|
"loss": 5.9141,
|
|
"mean_token_accuracy": 0.16307085007429123,
|
|
"num_tokens": 6108677.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.985226345062256,
|
|
"epoch": 2.891276321443919,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004472839880758419,
|
|
"loss": 5.8355,
|
|
"mean_token_accuracy": 0.1627019852399826,
|
|
"num_tokens": 6117151.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 6.153121757507324,
|
|
"epoch": 2.8955737000429735,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0004470699504663242,
|
|
"loss": 5.9386,
|
|
"mean_token_accuracy": 0.1510341614484787,
|
|
"num_tokens": 6127167.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.983470487594604,
|
|
"epoch": 2.899871078642028,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004468555371248104,
|
|
"loss": 5.8253,
|
|
"mean_token_accuracy": 0.16334982514381408,
|
|
"num_tokens": 6136487.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 6.00304102897644,
|
|
"epoch": 2.904168457241083,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0004466407485194937,
|
|
"loss": 5.9395,
|
|
"mean_token_accuracy": 0.15517849698662758,
|
|
"num_tokens": 6145334.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 6.022475290298462,
|
|
"epoch": 2.9084658358401376,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004464255851193864,
|
|
"loss": 5.848,
|
|
"mean_token_accuracy": 0.1620742604136467,
|
|
"num_tokens": 6155062.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 6.071009969711303,
|
|
"epoch": 2.912763214439192,
|
|
"grad_norm": 2.03125,
|
|
"learning_rate": 0.0004462100473943194,
|
|
"loss": 5.7993,
|
|
"mean_token_accuracy": 0.16524257585406305,
|
|
"num_tokens": 6164313.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.963624572753906,
|
|
"epoch": 2.9170605930382467,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000445994135814941,
|
|
"loss": 5.8595,
|
|
"mean_token_accuracy": 0.1599714934825897,
|
|
"num_tokens": 6173513.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 6.011693859100342,
|
|
"epoch": 2.9213579716373013,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00044577785085271566,
|
|
"loss": 5.8864,
|
|
"mean_token_accuracy": 0.1601871967315674,
|
|
"num_tokens": 6182000.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 6.079471397399902,
|
|
"epoch": 2.925655350236356,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004455611929799235,
|
|
"loss": 5.9065,
|
|
"mean_token_accuracy": 0.14938923865556716,
|
|
"num_tokens": 6191887.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.96527681350708,
|
|
"epoch": 2.9299527288354104,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004453441626696585,
|
|
"loss": 5.9729,
|
|
"mean_token_accuracy": 0.14858627021312715,
|
|
"num_tokens": 6202897.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 6.0613068580627445,
|
|
"epoch": 2.934250107434465,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00044512676039582823,
|
|
"loss": 5.8398,
|
|
"mean_token_accuracy": 0.16697479486465455,
|
|
"num_tokens": 6211811.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 6.068501138687134,
|
|
"epoch": 2.9385474860335195,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004449089866331524,
|
|
"loss": 5.8426,
|
|
"mean_token_accuracy": 0.16709824949502944,
|
|
"num_tokens": 6219896.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.907006549835205,
|
|
"epoch": 2.942844864632574,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004446908418571617,
|
|
"loss": 5.8423,
|
|
"mean_token_accuracy": 0.16006975322961808,
|
|
"num_tokens": 6228212.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 6.015281391143799,
|
|
"epoch": 2.9471422432316285,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004444723265441973,
|
|
"loss": 5.9968,
|
|
"mean_token_accuracy": 0.15501681417226792,
|
|
"num_tokens": 6238133.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 6.001640701293946,
|
|
"epoch": 2.9514396218306835,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004442534411714092,
|
|
"loss": 5.9186,
|
|
"mean_token_accuracy": 0.1549465298652649,
|
|
"num_tokens": 6247331.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 6.124386739730835,
|
|
"epoch": 2.955737000429738,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00044403418621675555,
|
|
"loss": 5.9002,
|
|
"mean_token_accuracy": 0.1585755541920662,
|
|
"num_tokens": 6255280.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 6.064080762863159,
|
|
"epoch": 2.9600343790287926,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004438145621590017,
|
|
"loss": 5.8598,
|
|
"mean_token_accuracy": 0.16812993735074996,
|
|
"num_tokens": 6264752.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.939321899414063,
|
|
"epoch": 2.964331757627847,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00044359456947771857,
|
|
"loss": 5.8138,
|
|
"mean_token_accuracy": 0.15746458172798156,
|
|
"num_tokens": 6273258.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.879025268554687,
|
|
"epoch": 2.9686291362269017,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004433742086532824,
|
|
"loss": 5.7338,
|
|
"mean_token_accuracy": 0.17725543826818466,
|
|
"num_tokens": 6281584.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.971160793304444,
|
|
"epoch": 2.9729265148259563,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00044315348016687317,
|
|
"loss": 5.8426,
|
|
"mean_token_accuracy": 0.16023122221231462,
|
|
"num_tokens": 6290016.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.993841552734375,
|
|
"epoch": 2.977223893425011,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004429323845004736,
|
|
"loss": 5.7544,
|
|
"mean_token_accuracy": 0.16837896257638932,
|
|
"num_tokens": 6298569.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.943963193893433,
|
|
"epoch": 2.9815212720240654,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00044271092213686824,
|
|
"loss": 5.8016,
|
|
"mean_token_accuracy": 0.16829360276460648,
|
|
"num_tokens": 6307684.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 6.045401191711425,
|
|
"epoch": 2.98581865062312,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00044248909355964247,
|
|
"loss": 5.9195,
|
|
"mean_token_accuracy": 0.1594064012169838,
|
|
"num_tokens": 6317767.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 6.1728271484375,
|
|
"epoch": 2.9901160292221745,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00044226689925318117,
|
|
"loss": 5.943,
|
|
"mean_token_accuracy": 0.15909445732831956,
|
|
"num_tokens": 6327457.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.942727422714233,
|
|
"epoch": 2.994413407821229,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00044204433970266785,
|
|
"loss": 5.7629,
|
|
"mean_token_accuracy": 0.17102586776018142,
|
|
"num_tokens": 6335747.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.950855731964111,
|
|
"epoch": 2.9987107864202835,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004418214153940837,
|
|
"loss": 5.8395,
|
|
"mean_token_accuracy": 0.16101130843162537,
|
|
"num_tokens": 6344750.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 6.046880510118273,
|
|
"epoch": 3.002578427159433,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00044159812681420624,
|
|
"loss": 5.8067,
|
|
"mean_token_accuracy": 0.16284387310345969,
|
|
"num_tokens": 6354779.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 6.0264809131622314,
|
|
"epoch": 3.0068758057584875,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004413744744506086,
|
|
"loss": 5.5785,
|
|
"mean_token_accuracy": 0.17180316895246506,
|
|
"num_tokens": 6363809.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 3.0068758057584875,
|
|
"eval_entropy": 5.832996326523858,
|
|
"eval_loss": 6.154844284057617,
|
|
"eval_mean_token_accuracy": 0.15930042531643365,
|
|
"eval_num_tokens": 6363809.0,
|
|
"eval_runtime": 2.6733,
|
|
"eval_samples_per_second": 1327.553,
|
|
"eval_steps_per_second": 166.084,
|
|
"step": 3500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 11630,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 9678727905408000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|